Sustainable Intermodal Supply of Biofuels Fawad Awais fawadawais@hotmail.com Licentiate thesis in Business Administration A P R I L 2 0 1 5 i Acknowledgments This learning endeavour would not have been possible without a few important indi- viduals and organisations. First of all, I would like to thank my supervisors, Jonas Flodén and Johan Woxenius. They not only gave me the opportunity for professional growth but, as role models, also influenced my personal growth. This work was funded by the Swedish Transport Administration (Trafikverket), Logistics and Transport Society (LTS) and Göteborg Energi under the project sustainable intermodal supply systems for biofuel and bulk freight, which have my deepest gratitude. Insights from professionals of other project members such as Skogforsk, BOKU, WSP, and Mariterm have been tremendously helpful. Above all, I am thankful to all the combined heating power plants in Sweden for providing their valuable data for the survey study. On a personal level, I cannot thank my colleagues enough. They not only made me feel at home in a foreign country but also helped me through bouts of professional and per- sonal turbulence. Some people I should recognise are Niklas Arvidsson, Zoi Nikopoulou, Viktor Elliot, Catrin Lammgård, and Taylan Mavruk. Lastly, I would like to thank God and my friends and family who kept my spirits high as I went through the various stages of life that were unknown to me. Mudassar, Salman, and Zohaib are true friends who were always there for me, in both thick and thin. My brother Ammar, my sister Jawaria, and my parents believed in me even when I did not believe in myself. The love and support of Sarah, my wife, have made the journey of life worth taking. My deepest apologies to the people who I forgot to mention here but who surely played a vital role in the completion of this work. Göteborg, on a cold and windy March day 2015. ii Sustainable Intermodal Supply of Biofuels Abstract Sweden shows significant consumption of forest fuels in its heating plants (HPs) and combined heat and power plants (CHPs) that places great demands on the logistics systems supplying these plants with fuel. The aim of this study is to help in the development of sus- tainable supply chains involving the intermodal bulk flow of wood as a fuel for Sweden’s HPs and CHPs. The study has involved an investigation of the definitions used for wood biofuels and their raw materials in the literature. After generally used distribution networks are identi- fied and analysed, the study describes the various logistical challenges in the wood biofuel industry. Information and data have been obtained from a literature review and survey, fol- lowed by a case study. Key challenges identified in the literature review are seasonal variations, storage, the chipping process, the low density of wood biofuels, the absence of standard terms, sources of supply, and dependency on policies. The survey reveals the situation of wood biofuel supply chains in Sweden. The key fuel used by the country’s power plants is woodchips, which has underscored their importance in keeping heat and electricity resources sustainable. The indus- try is oriented toward a local market that mostly uses trucks with direct transport of wood from the forest, the preferred site for chipping. Road transport is rated quite favourably, with reliability as the most important factor. At the same time, storage is used to overcome fluctu- ations in demand and is an essential part of the supply chain, as most CHPs have storage facil- ities. On this point, challenges include determining the size and location of storage facilities and identifying alternative possibilities for transport that might improve the transport chain and reduce environmental impacts, while at once maintaining flexibility. The case study ex- plores the sustainability of the various chains. The assessment of these chains considers costs and CO2 calculations. Frequent use and keeping transport distances short play important roles in keeping costs down. Large costs are associated with the terminal and chipping processes. All-road systems for wood biofuels often involve terminal costs which is their common char- acteristic with intermodal chains making the intermodal system potentially applicable due to low added costs. Keywords: Wood biofuel, transport, logistics, survey, logistical challenges, market, sustainability. iii Sammanfattning Sverige använder mycket skogsbränsle i fjärrvärmeverk. Detta ställer stora krav på lo- gistiksystemet som försörjer verken med bränsle. Denna studie syftar till bidra till utveckling- en av hållbara försörjningskällor med ett fokus intermodala flöden inom Sverige för fjärrvär- meverk. Studien inbegriper en litteraturgenomgång av definitioner för biobränslen och dess råmaterial. De vanligast förekommande distributionsnätverken identifieras och analyseras. Studien beskriver de olika logistiska utmaningarna i biobränsleindustrin. Data samlas in ge- nom en litteraturgenomgång och en enkät, följt av en fallstudie. Viktiga identifierade utmaningar genom litteratursstudien är säsongsvariationer, lag- ringen, flisningen, biobränsles låga densitet, avsaknaden av standardtermer, råvarukällor and beroendet av politiska beslut. Enkäten visar situationen för försörjningskedjorna för biobränsle i Sverige. Det huvudsakliga bränslet som används är träflis. Industrin har ett lokalt fokus och använder mest lastbilstransporter direkt från skogen. Lastbilstransporter rankas som det tydligt mest föredragna transportslaget, med tillförlitligheten som den viktigaste faktorn. Lagring används för att hantera variationer i efterfrågan och är en väsentlig del i försörjnings- kedjan, där de flesta verk har lagringsmöjligheter. Den mest föredragna platsen för flisningen är i skogen. I utmaningarna ingår att fastställa storleken och platsen för lagringen samt att identifiera alternativa transportmöjligheter som kan förbättra transportkedjan och leda till lägre miljömässig påverkan, samtidigt som flexibiliteten i kedjan behålls. Fallstudien under- söker hållbarheten i ett antal, existerande eller möjliga, försörjningskedjor som kan användas för att försörja verk i Sverige. Utvärderingen av kedjorna baseras på kostnad och CO2 utsläpp. Ett högt utnyttjande av resurserna och att hålla transportavstånden korta är viktigt för att hålla kostnaderna nere. Stora kostnader kan kopplas till terminaler och flisningen. Flera aktiviteter förekommer både i ett vägsystem och i ett intermodalt system vilket gynnar en övergång till ett intermodalt system. Lagernivåerna spelar en viktig roll vid beställningen av biobränsle. Nyckelord: Biobränsle, transport, logistik, enkät, logistiska utmaningar, marknad, hållbarhet. Author: Fawad Awais Language: English Pages: 95 Licentiate Thesis 2015 Department of Business Administration School of Business, Economics and Law University of Gothenburg P.O Box 610, SE 405 30 Göteborg, Sweden 1 Table of Contents Acknowledgments ....................................................................................................................... i Abstract ...................................................................................................................................... ii Sammanfattning ........................................................................................................................ iii Table of Contents ....................................................................................................................... 1 List of figures ............................................................................................................................. 4 List of tables ............................................................................................................................... 4 List of Appended Papers ............................................................................................................ 5 List of Abbreviations .................................................................................................................. 5 1. Introduction ......................................................................................................................... 6 1.1. Background ............................................................................................................................... 6 1.2. Discussion of the Problem ...................................................................................................... 10 1.3. Purpose ................................................................................................................................... 11 1.4. Research Questions ................................................................................................................. 12 1.5. Delimitations .......................................................................................................................... 12 1.6. Importance of the Study .......................................................................................................... 12 2. Frame of Reference ........................................................................................................... 14 2.1. Definitions of Biofuels ........................................................................................................... 14 2.2. Biofuel Supply Chains ............................................................................................................ 15 2.3. Industry Actors ....................................................................................................................... 16 2.3.1. The forest sector .............................................................................................................. 16 2.3.2. Wood processing industry ............................................................................................... 17 2.3.3. District heating ................................................................................................................ 18 2.4. Sustainability .......................................................................................................................... 19 2.4.1. Environmental sustainability ........................................................................................... 20 2.4.2. Economic sustainability .................................................................................................. 22 2.4.3. Social sustainability ......................................................................................................... 25 2.4.4. Applying sustainability .................................................................................................... 27 2.5. Sustainable Transport ............................................................................................................. 28 2.6. Intermodal Transport .............................................................................................................. 29 2.6.1. Definitions ....................................................................................................................... 29 2 2.7. Chain Components.................................................................................................................. 29 2.7.1. Terminals ......................................................................................................................... 30 2.7.2. Load units ........................................................................................................................ 31 2.7.3. Transportation ................................................................................................................. 32 2.7.4. Chain characteristics ........................................................................................................ 33 2.8. Swedish Railroad Intermodal Transport System .................................................................... 34 2.9. Wood Biofuel Supply Chain Actors ....................................................................................... 35 2.9.1. Terminals ......................................................................................................................... 35 2.9.2. Transport actors ............................................................................................................... 36 3. Methodology ..................................................................................................................... 38 3.1. Research Theme...................................................................................................................... 38 3.2. Data collection methods ......................................................................................................... 39 3.2.1. Literature review ............................................................................................................. 39 3.2.2. Survey study .................................................................................................................... 40 3.2.3. Case Study ....................................................................................................................... 41 3.3. Validity and reliability ............................................................................................................ 42 3.3.1. Validity ............................................................................................................................ 43 3.3.2. Reliability ........................................................................................................................ 45 4. Summary of appended papers ........................................................................................... 47 4.1. The appended papers in brief .................................................................................................. 47 4.1.1. Wood biofuels logistical challenges in Sweden .............................................................. 47 4.1.2. Logistic requirements and characteristics of the Swedish wood biofuel industry ........... 47 4.1.3. Meeting the challenges for intermodal transportation of biofuel .................................... 48 4.1.4. Project reports.................................................................................................................. 48 4.2. Logistics in wood biofuel transportation ................................................................................ 50 4.2.1. Harvesting and collecting biomass .................................................................................. 52 4.2.2. Storage ............................................................................................................................. 52 4.2.3. Transport in the bio-energy chain.................................................................................... 52 4.2.4. Pre-treatment techniques ................................................................................................. 53 4.3. Logistical Challenges.............................................................................................................. 53 4.3.1. Seasonal variations .......................................................................................................... 53 4.3.2. Storage ............................................................................................................................. 54 4.3.3. Chipping process ............................................................................................................. 54 4.3.4. Low density of wood biofuels ......................................................................................... 54 4.3.5. Term standardisation ....................................................................................................... 55 4.3.6. Sources of supply ............................................................................................................ 55 3 4.3.7. Dependence on policies ................................................................................................... 55 4.3.8. Logistical challenges identified from survey study ......................................................... 56 4.4. Swedish wood biofuel logistics .............................................................................................. 57 4.4.1. Operation 1: Harvesting and collection ........................................................................... 58 4.4.2. Operation 2: Storage ........................................................................................................ 58 4.4.3. Operation 3: Transport .................................................................................................... 59 4.4.4. Operation 4: Pre-treatment techniques ............................................................................ 64 4.4.5. Overall operation of the supply chain ............................................................................. 64 4.5. Case of Sävenäs Power plant .................................................................................................. 65 4.5.1. Case introduction ............................................................................................................. 66 4.5.2. Case methodology ........................................................................................................... 67 4.5.3. Break-even distance ........................................................................................................ 69 4.5.4. Base scenario ................................................................................................................... 70 4.6. Base scenario variations ......................................................................................................... 71 4.7. Best feasible case scenario ...................................................................................................... 75 4.8. Supply risk analysis ................................................................................................................ 76 4.8.1. One, two or three consecutive train deliveries missed .................................................... 79 4.8.2. Four or five consecutive trains missed ............................................................................ 79 4.8.3. Missing train analysis at the Sävenäs plant ..................................................................... 80 5. Conclusions and future research ....................................................................................... 82 5.1. Research questions answered ................................................................................................. 82 5.2. Logistics processes ................................................................................................................. 83 5.2.1. Operation 1: Harvesting and collection ........................................................................... 83 5.2.2. Operation 2: Storage ........................................................................................................ 83 5.2.3. Operation 3: Transport .................................................................................................... 83 5.2.4. Operation 4: Pre-treatment techniques ............................................................................ 84 5.2.5. Overall operation of the supply chain ............................................................................. 85 5.3. Challenges Explained ............................................................................................................. 85 5.3.1. Sustainability in wood biofuel supply chains .................................................................. 88 5.4. Future research ....................................................................................................................... 89 5.4.1. Sustainable international intermodal chains .................................................................... 89 5.4.2. Supply chain development based on fuel type ................................................................ 89 5.4.3. Development of business models .................................................................................... 89 5.4.4. Social sustainability ......................................................................................................... 90 5.4.5. GIS based study ............................................................................................................... 90 References ................................................................................................................................ 91 4 Appendices ............................................................................................................................... 97 Survey ............................................................................................................................................... 97 Cost Data ......................................................................................................................................... 110 Paper 1 ............................................................................................................................................ 111 Paper 2 ............................................................................................................................................ 127 Paper 3 ............................................................................................................................................ 151 List of figures FIGURE 1: DIFFERENT TYPES OF WOODCHIPS. ........................................................................................................ 15 FIGURE 2: THE BIOFUEL SUPPLY CHAIN ................................................................................................................. 16 FIGURE 3: THE RELATIONSHIP AMONG THE THREE PILLARS OF SUSTAINABILITY. .................................................. 20 FIGURE 4: A TYPICAL INTERMODAL SUPPLY CHAIN................................................................................................ 30 FIGURE 5: NATIONAL SUPPLY CHAINS UNDER FOCUS. ............................................................................................ 51 FIGURE 6: THE DIFFERENCES BETWEEN DIFFERENT TYPES OF BIOMASS ................................................................ 55 FIGURE 7: LOCAL TRACK LAYOUT AND ADJACENT SHUNTING YARD. .................................................................... 66 FIGURE 8: BREAK-EVEN ANALYSIS OF ROAD AND INTERMODAL SOLUTIONS. ......................................................... 69 FIGURE 9: THE PLANT AND SOURCING LOCATIONS.. ............................................................................................... 70 FIGURE 10: COSTS AND EMISSIONS IN THE BASE SCENARIO.................................................................................... 71 FIGURE 11: SUMMARY OF COSTS BASED ON VARIATIONS IN THE BASE SCENARIO. ................................................. 73 FIGURE 12: CHANGE IN COSTS AND EMISSIONS FROM THE BASE SCENARIO FOR TESTED CASES. ............................ 74 FIGURE 13: DISTRIBUTION OF COSTS AND CO2 EMISSIONS FOR THE BEST FEASIBLE CASE SCENARIO. .................... 76 FIGURE 14: DELIVERIES AND STORAGE LEVELS (EVENING) IN THE BASE SCENARIO. .............................................. 77 FIGURE 15: EXAMPLE OF POSSIBLE STORAGE AND DELIVERIES DURING ONE DAY. ................................................. 77 FIGURE 16: INCREASING COSTS BASED ON NUMBER OF WEEKS WITH 50% FULL TRAINS. ....................................... 78 List of tables TABLE 1: ENERGY PRODUCED BY DISTRICT HEATING PLANTS IN SWEDEN FROM DIFFERENT FUELS ......................... 9 TABLE 2: RESEARCH DESIGN.................................................................................................................................. 38 TABLE 3: VARIOUS STEPS OF SCM/LOGISTICS AND BIO-ENERGY. .......................................................................... 50 TABLE 4: MAIN LOADING AND UNLOADING TECHNIQUES OF WOOD BIOFUELS ....................................................... 53 TABLE 5: ISSUES IN THE SWEDISH BIOFUEL INDUSTRY ........................................................................................... 56 TABLE 6: MEAN RANKING OF STORAGE PROBLEMS. ............................................................................................... 56 TABLE 7: MEAN RANKING OF TRANSPORT PROBLEMS ........................................................................................... 57 TABLE 8: CHP SIZE. ............................................................................................................................................... 57 TABLE 9: FUEL USED. ............................................................................................................................................. 58 5 TABLE 10: SHARE OF RESPONDENTS HAVING STORAGE. ........................................................................................ 59 TABLE 11: AVERAGE STORAGE TIME IN DAYS. ....................................................................................................... 59 TABLE 12: TRANSPORT CHAINS USED. ................................................................................................................... 60 TABLE 13: TRANSPORT DISTANCES. ....................................................................................................................... 61 TABLE 14: RANKING OF IMPORTANT MODAL CHOICE FACTORS AND SERVICE RECEIVED ....................................... 62 TABLE 15: QUALITIES OF DIFFERENT TRANSPORT CHAINS. .................................................................................... 63 TABLE 16: MEAN RANKING OF PREFERRED TRANSPORT MODE. ............................................................................. 63 TABLE 17: LOAD UNITS/VEHICLES USED. ............................................................................................................... 64 TABLE 18: SHARE OF ENERGY PRODUCED BY VARIOUS CHIPPING LOCATIONS AND THEIR PREFERENCE ................. 64 TABLE 19: HANDLING FLUCTUATIONS IN DEMAND ................................................................................................ 65 TABLE 20: SERVICES RECEIVED AND THEIR IMPORTANCE ..................................................................................... 65 TABLE 21: ENERGY WHEN PLANT OPERATING AT MAXIMUM CAPACITY. ................................................................ 67 TABLE 22 COST LITERATURE SOURCES .................................................................................................................. 68 TABLE 23: NUMBER OF TRUCKS AND TRAINS NEEDED IN DIFFERENT SCENARIOS. .................................................. 80 TABLE 24: SUMMARY OF THE RESEARCH QUESTIONS ............................................................................................. 82 TABLE 25: LOGISTICAL ISSUES IN THE SCM STEPS ................................................................................................ 85 TABLE 26. LOGISTICAL CHALLENGES IDENTIFIED FROM THE SURVEY STUDY. ....................................................... 87 List of Appended Papers Paper 1 Awais, F., 2013, Wood biofuels logistical challenges in Sweden. Presented at the NOFOMA conference 2013, Gothenburg, Sweden. Paper 2 Awais, F. Flodén, J., 2013, Logistic requirements and characteristics of the Swedish wood biofuel industry. Submitted to the Scandinavian Journal of Forest Research. Paper 3 Flodén, J., Awais, F., 2014. Meeting the challenges for intermodal transportation of biofuel. List of Abbreviations DH: District heating HP: Heating plant CHP: Combined heating plants FAO: Food and agricultural organisation GHG: Greenhouse gases MW: Megawatt MWh: Megawatt hour GWh: Gigawatt hour TWh: Terawatt hour 6 1. Introduction This section provides an overview of the definitions of wood biofuels and different supply chains, along with a discussion of the problems. 1.1. Background The extent of economic and civil growth has always been associated with the con- sumption of natural energy resources (Mikkilä et al., 2009). At present, the growing need for energy resources poses numerous problems for the world, while the presence of oil and gas resources within only a handful of countries raises concerns of steady availability and the constant threat of their depletion. The use of fossil fuels has also been a chief contributor to environmental problems such as air pollution and Green House Gas emissions. Possible solu- tions to these problems call for the development of renewable, environmentally friendly ener- gy resources, among which the use of biomass presents an interesting alternative. Some of the many reasons for adopting biomass as an energy source include its worldwide availability, use in power generation, and the CO2-neutral basis of its biofuels (Hamelinck et al., 2005). Wood, an old and environmentally sustainable biofuel still used for energy purposes, is the focus of this study. Timilsina and Shrestha (2011) report that interest in biofuels as an alternative to fossil fuels emerged during the oil crisis in the 1970s. The subsequent price drops and incentives in the oil industry later stagnated the trend of biofuel production in many countries. Yet, with anticipated energy shortages in the coming years, along with increased oil prices and climate deterioration, interest in biofuels has been renewed. Its resurgence was further supported by the expansion in output and consumption of biofuels, along with ad- vancements in the technologies available. The decreased production cost of biofuels is a major motivator to use such means of energy instead of expensive oil sources. However, in nearly every case, biofuels still require subsidies to compete with oil products (e.g., gasoline and diesel). Climate preservation also dictates the use of biofuels against fossil fuels, given the former’s lesser effects on the climate via reduced CO2 emissions. Concerns such as rising oil prices and resource shortages call for further investigation of biofuels’ increased production and decreased production costs. According to the Swedish District Heating Association, the district-heating (DH) sec- tor has shown a steady reduction in the use of the fossil fuels since the 1980s. Currently, most of the energy supplied to Swedish heating plants (HPs) is renewable. These circumstances derive mostly from an elevated carbon tax among the industry’s recent measures to reduce the use of fossil fuels, which has indeed led to major reductions in carbon emissions (Trad, 2010). 7 Sweden uses biofuels in large quantities for the purposes of DH, combined heat, and electricity production. The use of forest fuels or woodchips as biofuel has increased over the previous decade and could continue to be a significant part of future fuel used. In 2010, Swe- den consumed 8.4 million m3 of forest chips for the purpose of energy generation. Logging residues form the most common raw material for the production of woodchips in Sweden (Routa et al., 2012). Obviously, this consumption has been possible due to the development of infrastructure necessary for the production of energy from biofuels. During the late 1970s, nearly 90% of the DH in Sweden was supported by oil, which later changed due to the oil crisis of the time. Numerous new HPs were built, while old plants were converted to accom- modate biomass. Commonly used fuels at HPs and combined heating plants (CHPs) are wood residues (e.g., from sawmills and forests), recovered wood, and refined wood fuels (e.g., wood pellets). During the 1990s, the support scheme for CHPs resulted in the construction of several large plants around Sweden. By 2000, nearly every town and city in Sweden had an HP or CHP using local biofuels. At present, Sweden has nearly 500 HPs around the country (Andersson, 2012). Table 1 shows that the consumption of fossil fuels (e.g., coal) has de- creased over time in the DH sector while the use of biofuels (e.g., wood biofuels) has risen. The increasing demand for energy has sharpened the focus on the logistics of supply- ing plants with fuel, since logistics is considered to pose key challenges for the increased use of biofuels (Gold and Seuring, 2011, Svanberg and Halldórsson, 2013, Rentizelas et al., 2009). The increased demand and production of wood biofuel calls for a closer examination of the logistics activities involved in transporting these goods to energy plants. The wood biofuel supply chain starts with the trees in the forest and ends with individual con- sumers. Along the way, it involves several processes: harvesting, sorting, transporting to ter- minals, along with sawmills, pulp mills, paper mills, and HPs, and the conversion of wood into products such as wood pellets, pulp, paper, and lumber (Carlsson and Rönnqvist, 2004). Sweden makes great use of these forest fuels in CHPs and HPs, which require different amounts of different wood biofuels. The DH sector uses nearly half of all biofuels consumed in Sweden and major developments in the DH sector during the past three decades. The DH sector serves almost half of Sweden’s population, including both commercial and residential buildings. Of late, HPs have been combined with the production of electricity, thus giving rise to CHPs. The high environmental tax on the use of fossil fuels in Sweden is a major reason for the DH sector’s shift from oil to biofuels. In fact, biofuels have replaced oil as a source of fuel in most places, which has resulted in the tremendous demand for biofuels by 8 HPs. Another contributing factor is Sweden’s endowment and enrichment of forests, which are major sources of wood used as biofuel (Olsson, 2006). The growing demand for bioenergy requires the long-distance transport of wood. This necessity highlights the importance of rail in the transport of both biofuels and traditional for- est products (Tahvanainen and Anttila, 2011). Long-distance transport involving intermodal modes can reduce costs and involve the transport of large volumes to meet demand. The transport of wood biofuels presents interesting scenarios of intermodal transportation from the source to the destination, while the consumers (i.e., HPs and CHPs) look for efficient supply chains within Sweden. Sustainability is at the heart of using biofuels. The use of biofuels in the DH sector is a sus- tainable activity that greatly reduces emissions. It can be argued that the supply of wood bio- fuels causes the most emissions in the whole process of their use, since the supply of wood biofuels relies heavily on processes that use fossil fuels and assumes variable costs, making it the least sustainable part of the whole chain. The great dependency on road transportation in wood biofuel supply chains negatively affects the whole process given the high costs associ- ated with long-distance transport. In response, a focus on the sustainability of wood biofuels would greatly contribute to improving the sustainability of the total process. Sustainability in wood biofuel supply chains can be seen as a step toward progress in an otherwise highly sus- tainable process. As such, this thesis aims to help in the development of sustainable supply chains involving the intermodal bulk flow of wood fuel among Sweden’s HPs and CHPs 9 Table 1: Energy produced by district heating plants in Sweden from different fuels (excluding electric- ity) 2006–2011, GWh. Source: Swedish District Heating Association (2011) Fuel 2011 2010 2009 2008 2007 2006 Industrial waste heat 3,852.3 4,121.5 3,589.8 3,842.2 3,739.9 3,785.1 Solar n/a n/a 8.1 Waste 9,581.4 10,191.1 9,477.7 7,719.7 7,285.6 7,458.8 Waste gas 718.7 740.4 574.4 870.1 844.3 828.9 Recycled woodchips 2,445.1 2,906.5 3,165.5 2,338.8 1,453.7 1,321.0 Logging residues, stem woodchips, sawdust, etc. 14,284.4 18,765.1 16,716.7 13,642.0 11,823.1 14,182.6 Pellets, wood briquettes, etc. 3,470.6 4,579.8 4,012.0 4,023.0 3,479.2 3,882.9 Landfill and sewage gas 115.5 173.4 128.6 129.1 26.1 245.9 Tall oil pitch 710.8 983.7 862.5 737.9 667.7 743.7 Bio oil 960.0 2,256.4 2,072.7 1,309.3 1,641.7 1,713.7 Other biofuel 3,288.1 3,498.6 788.9 Other fuel 840.0 783.4 995.0 Peat and peat briquettes 1,726.9 2,674.3 2,608.0 2,549.3 2,583.7 2,166.6 Purchased hot water (unspecified fuel) 140.4 17.3 47.5 183.0 600.5 652.4 Electricity for heat pumps 1,264.6 1,427.8 1,436.6 1,564.6 1,643.1 1,553.9 Heat output from heat 3,921.3 4,574.5 4,659.7 4,768.2 5,164.4 5,064.2 Electricity for electric boilers 144.7 139.4 211.1 221.3 339.3 235.9 Support electricity 1,758.4 1,792.1 1,458.8 1,382.1 1,612.0 1,156.1 Natural gas 2 ,11.1 3,306.0 2,451.9 1,675.8 2,049.1 1,721.6 Heating oil 2,068.9 4,558.1 3,836.7 1,269.8 1,686.4 2,701.9 Coal 1,347.6 1,606.1 1,443.7 1,449.4 1,803.0 1,947.4 Other fossil fuel 171.8 325.1 498.1 229.0 323.9 265.4 Flue gas condensation 3,899.0 Total fuel / energy for heat 53,428.9 63,710.6 57,815.5 52,468.1 51,405.6 51,865.9 Total heat supply 48,079.5 61,171.9 50,825.1 47,758.6 47,432.4 46,735.9 Efficiency 97% 96% 88% 91% 92% 90% 10 1.2. Discussion of the Problem This study investigates the supply chain of wood used for energy from a holistic per- spective. The discussion of the problem therefore uses general terms to address the need of biofuels and logistical aspects of the wood biofuel industry. Climate preservation is at the heart of the concept of forest fuels. With growing de- mand for forest fuels, improving supply networks is a logical possible next step for the DH sector. Rauch and Gronalt (2010) explain that, logistically, wood is heavy to transport and provides less energy than fossil fuels, which suggests an economic dilemma in the transport of forest fuels. Since the transport of wood fuels from widespread sources to widespread destina- tions involves far higher costs than fossil fuels, cost-effective solutions should be sought. Several factors can be evaluated for ways to make wood transport more cost-effective, includ- ing the mode of transport, physical condition (e.g., chipped, unchipped, baled) of the wood, and moisture content, among others. The main cost drivers in developing a supply chain of wood fuels are chipping and storage, while present and future energy costs are forecast to remain high or increase. Continually increased demand and consumption, along with dynamic aspects of sup- ply chains, of wood biofuels present an opportunity to study the phenomenon in detail. With its substantial consumption and large number of HPs and CHPs using biofuels, Sweden is an ideal place for studying the supply chains of wood biomass. CHPs and HPs require fuels in large quantities and facilitate the study of different aspects of intermodal transport. Large CHPs and HPs with significant consumption can withstand costs related to intermodal transport options, including the combination of rail and road transport, by increasing resource use. As such, CHP operations in Sweden present an opportunity to study the situation of the wood biofuel industry and define the logistical problems and sustainability of the wood biofu- el supply chain. Aside from the necessity of both biofuels and improving the wood biofuel supply sys- tem in the today’s energy-deprived world, presented research has aspired flesh out the study of logistics and transport management. In this sense, wood biofuel supply systems present opportunities to study key concepts in intermodal transport, including sustainable transport, storage, terminal, and capacity management, among many others. In the current study, focus was kept on sustainable intermodal transport. Priemus et al. (1999) describe freight transportation in Europe, which has increased compared with passenger transport. Intermodal transportation can be used to reduce traffic on 11 the road, which will prevent current and possible congestion problems. Freight transportation most affects the environment when it uses trucks. Greener approaches to freight transportation include the use of pipelines, ships, and trains. To increase the use of environmentally friendly modes, terminals with advanced transhipment techniques should be developed in the right places. At the same time, advanced terminals and networks posing reduced costs and envi- ronmental impacts can be achieved by better configurations of mode combination, terminals, and freight flow. Wiegmans et al. (1999) highlight the importance of handling operations at a terminal, which are thought to constitute an expensive part of the supply chain. In this sense, the objective should be reduced operations. Modifying handling operations is acceptable only if given a significant increase in the performance of the terminal, a reduction of costs, or a combination of both. Jourquin et al. (1999) further reinforces this point by stating that im- provements and innovations introduced by the use intermodal transport should be both feasi- ble and economical in order to pose benefits. The supply chain of wood biofuels can benefit from intermodality since it involves the use of storage and transhipment terminals. Multiple sourcing points can bring goods to termi- nal points, from which they can be transported in huge volumes to single destination points (e.g., power plants). Such a setup can be ideal for implementing intermodal solutions. With the growing demand for wood biofuels and the increase in overall freight transportation, shift- ing loads to different modes is a possible solution. Regardless, intermodal solutions involving trains and trucks need to be both feasible and economical. Wood biofuel supply chains in Sweden provide excellent grounds for studying both existing and potential intermodal solu- tions for products that contribute to the sustainability of society. 1.3. Purpose The purpose of this study is to investigate wood biofuel supply chains in Sweden and to facilitate the use of intermodal freight transport for supplying district heating plants. The study examines the various logistical activities and problems involved in the sup- ply chains of raw materials for HPs and CHPs in Sweden. These activities and problems are analysed in a context of sustainability to suggest sustainable transport chains for the supply of wood biofuels. More specifically, this thesis investigates the transportation of wood biofuels in multimodal transport systems, which currently dominate wood biofuel supply chains. In this sense, the thesis examines the potential of implementing intermodal transport systems within current chains. The benefits of using road and rail transport have been investigated in wood biofuel chains, as such taking advantage of the benefits of different modes is a top goal 12 of intermodal transport. The survey has largely focussed on current and possible intermodal activities of wood biofuel chains, while the case study lastly investigates the potential of in- termodal activities in wood biofuel supply chains. This thesis’ focus on the combination of road and rail transportation is highly motivat- ed by the present infrastructure in Sweden for rail–road intermodal transport. Rail and road transport are the most commonly used modes for transporting biofuels in Sweden, whereas ships are used only for imports. Section 2 highlights the Swedish rail–road intermodal system and briefly introduces how biofuels are currently transported in intermodal settings. 1.4. Research Questions To fulfil the purpose of this study, the research questions developed will be set as guiding beacons to identify and develop a sustainable supply chain for the wood biofuels. RQ 1: What are the different actors and practices involved in wood biofuel supply sys- tems for heating plants? RQ 2: What are the main preferences, requirements, and logistical challenges in the wood biofuel supply system for heating plants? RQ 3: How can sustainable intermodal transport options be designed for a wood bio- fuel supply system for heating plants? 1.5. Delimitations Despite various biofuels, this study is particularly concerned with wood to be used as biofuel. The raw material to be discussed for the selected product is wood biomass available from both wood processing plants and forests. Other applications of wood such as for furni- ture or other purposes are not the focus of this study, thus those applications’ various aspects and supply chains are excluded. The different modes of transportation discussed in the study are road and rail. The study is limited to Sweden’s HPs and CHPs that involve national supply chains. As one of the largest consumers of wood biofuels and showing increased usage, Swe- den provides an ideal situation for studying supply chains. 1.6. Importance of the Study The study is important from multiple points of view. Identifying logistical problems in wood biofuel chains provides potential starting points for improvement. The survey of CHPs highlights the current market situation along with attitudes toward different modes and logis- tics activities. The calculation of costs and CO2 emissions for a wood supply chain based on a case study highlight the economics and various options for sustainable logistics practices. The 13 study is most important from a logistics point of view, though provides insights into market trends and attitudes currently present among wood biofuel consumers. 14 2. Frame of Reference This section provides an overview of the background knowledge used in conducting the study. 2.1. Definitions of Biofuels The category of biofuels involves a diverse range of products and substances used to generate energy. Wood from trees is one of the most widely used solid biofuels and can be transformed into products such as wood pellets and torrefied wood (Bradley et al., 2009b). Compared to fossil fuels, however, wood is heavy and yields less energy (Rauch and Gronalt, 2010). The Food and Agricultural Organization (FAO) of the United Nations has developed common terms for biofuels, thereby providing a structured way to classify the various biofu- els available. The purpose of developing bioenergy terminology was to standardise definitions of various terms related to bioenergy for international usage. Differences in definitions due to local alterations have posed several problems that frustrated the comparison and report of dif- ferent regions. In response, Thraen et al. (2004) have focused on unifying and organising terms and definitions of wood and other biofuels used in forest and energy statistics, bioener- gy balances, and commercial trading operations. Woodchips used as fuels consist of a mixture of hard and soft woods reduced to a size of roughly 5–8 cm and heterogeneous in shape. Woodchips are classified according to mois- ture content, bulk density, net calorific value, energy density, and particle size. The UN FAO uses woodchips to refer to any chipped biomass mechanically reduced to a defined particle size. Mechanically processing of woodchips involves the use of sharp tools. The chipped biomass is usually rectangular in shape, 5–50 mm in length, and of a generally low thickness compared to its other dimensions. 15 Figure 1: Different types of woodchips (photo: Flodén). Woodchips come in many varieties. Cutter chips, either with or without bark, are woodchips produced as a by-product of the wood processing industry. Forest chips are chips of various forest woods in three subcategories: green chips, stem woodchips, and whole-tree woodchips. While green chips are made of fresh logging and thinning by-products, including branches and treetops, stem woodchips are made of trunk wood (i.e., the tree trunk without branches) and can be with or without bark. Lastly, whole-tree chips are made up of all tree parts, including trunks, bark, branches, needles, and leaves. Among other by-products related to woodchips are logging residues, which are wood biomass produced when merchandisable timber is harvested in forests. Logging residues derive from treetops and branches cut while fresh or after seasoning. 2.2. Biofuel Supply Chains Supply chains for forest fuels either deliver products directly to power plants or use terminals as buffers. Determining terminal locations and the various costs involved, along with the demand of wood fuels, presents complex scenarios for the industry in developing cost-effective, CO2-neutral energy (Rauch and Gronalt, 2010). The biofuel supply chain starts with natural forests and the wood processing industry (e.g., sawmills, the paper and pulp in- dustry), which are the main sources of raw materials necessary for the production of wood- chips. Figure 2 describes the general flow of wood products involving the interaction of im- portant actors in the wood industry. 16 Figure 2: The biofuel supply chain (Energidata AS et al., 2005). The primary operations of a normal biomass chain are harvesting and collection, stor- age, transport, and pre-treatment. Gold and Seuring (2011) reveal in their literature review that the overall design of biomass chains is the area most focused upon, followed by harvest- ing and collection. Topics with less scientific focus are biomass storage and pre-treatment techniques. Most of the literature focuses on the overall layouts of biomass supply chains, not their components. The area first focused upon in the biomass supply chain is supply chain architec- ture, which involves optimisation solutions for the location of storage and the chipping pro- cess. Biofuel chains are complex and involve different actors and market segments. Though energy plants that use biofuels are smaller than those using fossil fuels, their logistics is more complex, since they require more deliveries due to wood’s low energy content (Gold and Seuring, 2011). 2.3. Industry Actors Important industry actors in wood biofuel supply chains are discussed in this section, along with their current statuses in Sweden. 2.3.1. The forest sector The use of forest resources has generally always contributed significantly to the Swe- dish economy, and Sweden’s forestland is considered to be an economic resource. In support, for more than a century Swedish regulations have ensured the long-term productivity of for- ests. In 2000, Swedish exports from forests represented 13% of total exports, which demon- strates the sector’s economic importance to the country. In 1993, changes to Swedish regula- 17 tions regarding forestland gave both associated economic and environmental factors equal importance (Ericsson et al., 2004). The Swedish National Board of Forestry (2001) has determined that nearly half of Swedish forestland is owned by private, non-industrial owners, that companies own the other half, and that the Swedish state owns a very small share of forestland. However, the govern- ment has greater influence in the forest industry than private interests, since one third of the forestland is owned by companies managed by a government-owned company. The remaining percentage of the land is owned by other public organisations such as municipalities and other combined entities. Recent figures from the Swedish Forest Agency regarding ownership show the following breakdown: x 50% individual ownership; x 25% privately owned company ownership; x 14% state-owned companies ownership; x 6% other private ownership; x 3% state ownership; and x 2% other public ownership (Eriksson, 2011). An estimated 344,000 private forest owners in Sweden belong to one of three forest owners’ associations in the country. Forest owners’ associations provide assistance to forest owners with managing various forest operations (e.g. harvesting, sales) and inform their members of the importance of harvesting logging residues. Meanwhile, the three largest forest companies in Sweden are Stora Enso, SCA, and Södra. In general, the presence of organised forest owners in the Swedish forest sector indicates the strong influence of these actors on the sector. Other organisations that influence the sector are manufacturers of forest harvesting equipment and transport companies. Transport companies have played a particularly valuable role in the development of wood biofuels, for they could provide the existing transport infra- structure for wood biomass (Ericsson et al., 2004). 2.3.2. Wood processing industry The wood processing industry, including the pulp and paper industry, wields signifi- cant control over the flows of wood biomass, for they both are major consumers of wood and produce a large share of the raw materials for wood biofuels. The relationship between the forest and energy industry is also historically significant. Wood processing companies are often large buyers of electricity and own facilities that generate electricity with steam- powered turbines usually present at sawmills. In turn, sawmills are often involved in provid- 18 ing waste heat and electricity to neighbouring communities in addition to wood biofuels. Some pulp and sawmills have integrated into their operations the production of refined wood biofuels such as wood pellets. Swedish wood processing companies have also been involved in research and devel- opment programs for the industry, which has induced the coordinated harvesting of timber and logging residues for use as biofuel (Ericsson et al., 2004). Woodchips are also used as a raw material for the production of pulp and paper, though the quality requirements for chips to be used as pulp are higher than they are for energy use, making woodchips for pulp more expensive. A certain competition among the industries for raw materials exists, since power plants can easily burn high-quality pulp chips if the price is right. 2.3.3. District heating Like most northern European countries, building the DH sector in Sweden started with municipality initiatives and were later managed by municipalities as well. Later, control was shifted to municipally owned companies, some of which were later sold to large international utilities. These large companies now provide 42% of the energy produced by the DH sector. The Swedish population has generally accepted the DH system, which has supported its de- velopment (Ericsson, 2009). DH systems have been widely accepted in Sweden due to the country’s cold climate and thus seasonal high demand for heat energy among the general population. The main driv- ers behind the success of DH systems are high fuel efficiency, low emissions, and fuel flexi- bility compared to the single household heating system. Local authorities play a vital role in the physical planning and selection of heating systems in Sweden. Decisions regarding the development of infrastructure, including place of construction and type of both heating system and roads, are made by local authorities. In this regard, political decisions and the fixed costs of establishing a DH system are crucial (Ericsson et al., 2004). HPs require large investments, though their benefits have promoted their construction since the 1950s. The first 10 HPs in Sweden involved oil-powered CHPs. Later, with the development of the nuclear energy sector and lower electricity prices, the DH sector became less attractive. Yet, with the help of a Swedish scheme for tradable renewable electricity certificates introduced in 2003, invest- ments in the DH sector have returned. In 2007, the DH sector provided 7.5 TWh of energy, 42% of which was produced from biomass (Ericsson, 2009). Generating electricity is highly integrated into DH systems, which has given rise to CHPs. In Sweden, however, the potential of CHPs has not yet been fulfilled. A factor consid- 19 ered to hinder the developments of CHPs in Sweden is the dominance of nuclear energy in the electricity sector, which limits the economic growth of CHPs. The increased generation of nuclear electricity resulted in surplus electricity in the 1990s in Sweden, and nuclear power has been largely dominant ever since (Ericsson et al., 2004). 2.4. Sustainability Sustainability suggest that our economic systems should be managed in ways that al- low societies to live off of the dividends of current resources so that future generations will be able to live as well, if not better. The most common definition of sustainability comes from the Brundtland Commission, which defines sustainable development as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (World Commission on Environment and Development, 1987). This defini- tion highlights the concept of the needs of both present and future generations and how they should be aligned with nature’s ability to provide resources. Anand and Sen (2000) provide a more general discussion about sustainability and our responsibility toward future generations. Many other definitions of sustainability have been suggested in logistics research (e.g., Janic, (2006); Seuring and Müller, (2008); Carter and Easton (2011)). Sustainable development can further be divided into three areas: environmental sus- tainability, economic sustainability, and social sustainability (Carter and Rogers, 2008). While environmental sustainability concerns emissions and the use of natural resources, economic sustainability concerns the long-term profitability and survival of the system. Lastly, social sustainability concerns society and social responsibility, including aspects such as health and safety, employment, social equity, and human rights. Together, the three areas highlight the holistic view of sustainability, since all areas must be sustainable for the entire system to be sustainable. As Petter Stordalen has phrased it, ‘There is no business on a dead planet’.1 Sus- tainability also involves inter-community aspects, since a high level of sustainability in one area cannot outweigh unsustainability in another area. For example, it is not considered sus- tainable for a company to be green and have a very low environmental impact while taking economic losses and ultimately going bankrupt. 1 http://www.stordalenfoundation.no/EN/ 20 Figure 3: The relationship among the three pillars of sustainability (Carter and Rogers, 2008). Figure 3 illustrates the balance among environmental, social, and economic perfor- mance that sustainability seeks in any process. 2.4.1. Environmental sustainability Environmental sustainability refers to the maintenance of natural capital, which Goodland (1995) defines in terms of output and input rules. Whereas the output rule is related to waste emissions, which should be kept within the assimilative limits of an environment, the input rule relates to the use of renewables and non-renewables along with operational princi- ples. Sutton (2004 pp. 11) describes environmental sustainability as ‘the ability to maintain things or qualities that are valued in the physical environment’. Here, the physical environ- ment is the natural and biological environment around us. By some contrast, Park (2007) describes environmental sustainability as ‘the long- term maintenance of ecosystems and other environmental systems for the benefit of future generations’. From this definition, it is clear that environment sustainability results in the maintenance of natural resources in the state in which they exist, as well as benefits future generations, which is consistent with the Brundtland Report. By still greater contrast, Ekins (2011pp. 637) defines the term as the ‘maintenance of important environmental functions, and hence the maintenance of the capacity of the capital stock to provide those functions’. This definition associates the maintenance of capital with 21 the maintenance of vital environmental functions, which helps to explain environmental sus- tainability in economic terms. To achieve environmental sustainability, logistics is paramount, given the emissions produced by various processes in the supply chain. The environmental sustainability of a sup- ply chain, which is often referred to as the greening of the supply chain, involves environmen- tal issues associated with decisions in transport, storage, inventory control, warehousing, packaging, and facility location. The aim of green supply chain management is to reduce the carbon footprint of all activities involved in the chain (Min and Kim, 2012). Wiedmann and Minx (2008 pp. 4) define carbon footprint as ‘a measure of the exclusive total amount of car- bon dioxide emissions that is directly and indirectly caused by an activity or is accumulated over the life stages of a product’. CO2 released from vehicles is the most significant repre- sentative of air pollution, and its measurement provides meaningful information for develop- ing environment-related policies (Lumbreras et al., 2013). Companies willing to improve en- vironmental sustainability prefer that their suppliers reduce their environmental liability as much as possible (Sarkis, 1995). Assessing environmental sustainability. The definitions presented here all prioritise the maintenance of the environment in terms of processes and the state. Maintaining the envi- ronment requires monitoring the environment. Ekins (2011) discusses how environmental sustainability can be expressed in terms of capital, yet concludes that the valuation of envi- ronmental functions is extremely complex, given their non-marginal nature and/or high costs that arise from the loss of environmental functions. Defining safe minimum standards for the valuation of environmental functions has been proposed to reduce such complexity, and vari- ous challenges involved in developing these standards have been data generation and under- standing ecosystems. In past decades, these challenges were met with developments in cli- mate science and GHG accounting protocols. The OECD (2001) has outlined a strategy for achieving environmental sustainability, which involves five general objectives: x Maintaining the integrity of ecosystems via the efficient management of natural re- sources; x Decoupling environmental pressures from economic growth; x Improving decision-making processes by advancing the measurement process; x Enhancing quality of life; and x Achieving global environmental interdependence involving the improvement of gov- ernance and cooperation. 22 After certain indicators have been established for indicating sustainability, they should be measured both quantitatively and qualitatively. A chief difficulty in this regard is the selec- tion of appropriate indicators, not data collection. The different indicators may refer to differ- ent objectives or values depending upon the national or international policies (Moldan et al., 2012). Since environmental sustainability can be dictated by objectives, a prominent example in this regard is the European Council’s objective of reducing GHG emissions by 20%, mak- ing renewable sources 20% of all energy sources, and increasing energy efficiency by 20%— all by 2020 (EC, 2007). Such an understanding of environmental sustainability precipitates the development of measures to develop standards for monitoring the achievement of envi- ronmental objectives. In supply chains, environmental sustainability can be estimated by calculating the ex- ternal costs related to the chain. Button (1993) states that external costs occur when the wel- fare of one group is affected by the activities of another without any compensation. External effects can be either negative or positive, though most related to transportation are negative (e.g., noise, visual intrusion, risk of accident, emissions, congestion). External costs are also closely related to social costs and their valuation. External costs associated with freight trans- portation can be measured in different ways, though are always associated with the measure- ment of emissions (Mckinnon et al., 2012). The payment of external costs should not be con- sidered a sustainable activity, since it involves paying for an activity that can be avoided, meaning that the continuation of the practice would weaken economic sustainability. Such a practice would also undermine environmental sustainability, whose underlying principle is to conserve the environment for future generations and not to simply pay extra to be allowed to damage it. 2.4.2. Economic sustainability Barbier (1987) has outlined four criteria for sustainable economic development, which are as follows: 1. Economic growth cannot be separated from the society, since economic changes are associated with social, cultural, and ecological changes; 2. Any quantitative aspect of sustainable economic development is associated with the increase in materials for the present and future generations that live in poverty, which can substantially support physical and social well-being in efforts against poverty; 3. Any qualitative aspect is multidimensional and requires ensuring long-term ecological, social, and cultural potential in support of economic activity and structure; and 23 4. Quantitative and qualitative aspects are not easily measureable. The abovementioned criteria associate sustainable economic development with the in- crease of material standards for the poor. This material increase can be measured in terms of increased food, real income, educational services, healthcare, sanitation, and water supply, among other aspects. In sum, the objective of sustainable economic development is to reduce poverty by providing long-lasting livelihood while minimising resource depletion and envi- ronmental, cultural, and social damage (Barbier, 1987). Aspects of sustainable economic de- velopment can be associated with companies as well, which aid economic development by providing jobs, income, and other benefits to people in the society. While differentiating the three pillars of sustainability, Goodland (1995) has defined economic sustainability as the maintenance of capital that keeps economic capital stable. Eco- nomic sustainability can also be seen in terms of a firm’s social responsibility. Carroll (1979) states that principal social responsibility of a firm is to fulfil its economic responsibilities, which include making the organisation to act as a business in society and to produce goods to be sold at a profit. All other business roles should be based on this fundamental assumption. Moldan et al. (2012) stress that the economic crises have highlighted the need for economic sustainability; these crises urge countries to keep focus on the maintenance or restoration of economic capital. Nevertheless, striking a balance between economic growth and sustainabil- ity has been a challenge for modern societies. Carter and Rogers (2008) state that the economic responsibility of firms seems to be lacking in literature addressing logistics and purchasing social responsibility, which makes it difficult to define sustainable logistical practices. They argue that being environmentally and socially sustainable may or may not be profitable at times. Earlier, Bowen et al. (2001) de- scribe that green (i.e., environmentally sustainable) supply chain management practices would be adopted by organisations if particular financial or operational gains were involved. These financial gains can be regarded to contribute to the economic balance that organisations seek along with environmentally friendly activities. Rao and Holt (2005) have concluded that mak- ing supply chains greener has the same potential in terms of economic performance and com- petitiveness as that of non-green supply chains. They argue that if organisations implement green supply chains, then they will not only save costs but also be able to enhance sales, mar- ket share, and new market opportunities, thereby prompting enhanced economic performance. 24 Green activities can be referred to as economically sustainable activities involving the intersection of environmental, societal, social, and economic bottom lines. Carter and Rogers (2008) provide the following examples based upon their literature review: x Cost savings due to reductions in waste and packaging; x Reduced health and safety costs; x Reduced labour costs due to increased motivation with improved working conditions; x Proactively shaped future regulations by continuously focusing on changing environ- mental and social concerns; x Reduced lead times and costs result in better product quality; and x Improved reputation. From all of the above, it is clear that sustainable supply chain management involves the long-term economic performance of supply chains, not only their environmental and so- cial sustainability. Assessing economic sustainability. In the current study, economic sustainability is viewed according to Goodland (1995), who assumes it to be the maintenance of capital. Carroll (1979) view that any company’s first responsibility—to make profit—also fits well in supply chains in which costs should not outweigh profits. The examples of green activities given by Carter and Rogers (2008) said to be environmentally and economically sustainable involve mostly cost-saving activities. Therefore, to evaluate the economic performance of a process, its costs need to be identified. Long-term economic performance can be measured in terms of costs incurred by the various activities in a supply chain. In all, the estimation of costs and measurement of economic performance of a process contribute to the measurement of economic sustainability. Savitz and Weber (2006) definition of sustainability that refers to good corporate citi- zenship as a principle of smart management further reinforces Carroll (1979) ideal of making profit by reducing costs. To have smart management and economic sustainability in supply chains, a logical starting point is the estimation of the chain’s costs. Economic sustainability can be expressed in monetary terms easily divided into costs and revenue. The estimation of costs can help to define the economic performance of a firm or supply chain. Flodén (2007) explains that identifying the various costs in a supply chain can be complex. Identifying incomplete costs or revenues of a supply chain can lead to an incorrect representation of economic sustainability. Calculations of costs involve a great variety of es- timations since different factors and actors influence costs differently. While estimating costs, 25 determining the cost variables can be more helpful than the actual costs, since determining cost variables and what they depend upon allows the calculation of actual costs in different scenarios. The cost variables for a supply chain system have to relate to operational transport activities. The price and cost of transportation are two different concepts since the price charged for transportation can be influenced by a wide variety of factors. Like any industrial costs, transport costs can be divided into the two general categories of fixed and variable costs. These fixed and variable costs also depend upon the period in which they are studied. For example, the rent of a terminal area can be regarded as a fixed cost; however, if closing the terminal is an option, then rent can be regarded as a variable cost. Similarly, a predefined train schedule can be regarded as a fixed cost; however, if variability is possible in the schedule, then it should be deemed a variable cost. As such, fixed costs can be variable costs considered to be fixed for a specific period. Fixed costs can be further divided into other categories, including shared costs, which are shared by different actors or transport modes within a supply chain. Numerous studies show that variable costs can also be divided into two major categories: time and distance transported. Examples of common time dependent costs are financial costs, salary costs, vehi- cle taxes, and insurance, while common examples of distance-dependent costs are tires, fuels, maintenance, kilometre taxes, and rail infrastructure fees. It should be kept in mind that some of these costs can be fixed depending upon the time frame (Flodén (2007). 2.4.3. Social sustainability Social sustainability is an overlapping concept that involves topics such as social capi- tal, social cohesion, social inclusion, and social exclusion. Invariably, social sustainability is considered in light of the goals of social development, which can be highly diverse (Hopwood et al., 2005, Littig and Grießler, 2005). Similar to that of sustainability itself, the definition of social sustainability cannot be fixed, since it is a dynamic concept that changes over time. Changes may be caused by external influences; for example, changes in the local authority service can affect social cohesion and the interaction of societies (Dempsey et al., 2011). Labuschagne et al. (2005) have identified criteria for social sustainability for industries that encompass general themes of internal human resources, external population and stakeholder participation, and macro social performance. The criteria have been developed based upon various frameworks of sustainability such as the Global Reporting Initiative, the UN Commis- sion on Sustainable Development Framework, the Sustainability Metrics of the Institution of 26 Chemical Engineers, and Wuppertal Sustainability Indicators. The general themes given by Labuschagne et al. (2005) are: x Internal human resources, which focuses on the social responsibility of the company toward its employees; x External population, which is related to the external effects of the company on the community in which it operates; x Stakeholder participation, which is related to treating its stakeholders in an ethically and socially responsible manner. x Macro social performance, which involves the impact of a company on a regional or national level (Labuschagne et al., 2005). Hutchins and Sutherland (2008) describe how corporate social responsibility (CSR) is important to the social pillar of sustainability. Many definitions of CSR refer to ethical behav- iour related to the environment, society, and the economy. The goals of social sustainability and CSR tend to cater to basic needs by reducing poverty, increasing human health, and pro- tecting ecosystems, yet also include higher-level needs, as well such as education and gender equity. Companies can meet their social sustainability and CSR goals by fulfilling the basic needs of the society in which they exist and extending their efforts to meet higher-level needs such as safety, quality of life, and equity. Hutchins and Sutherland (2008) have also proposed four indicators of social sustaina- bility in the various actors of the supply chain that need evaluation: labour equity, health care, safety, and philanthropy. These indicators form portions of the criteria suggested by Labuschagne et al. (2005), which can be used to promote the social sustainability of the whole company. Lammgård (2007) argues that information about attitudes, preferences, and needs of the companies toward environmental or intermodal road–rail transport solutions will help to not only segment the industry but also to define the demands and needs of freight transporta- tion. Such findings can be used to clarify the acceptability of sustainable practices in supply chains by identifying the attitudes, preferences, demands, and needs of the different actors. The attitudes and preferences will also highlight the conditions in which sustainable activities can be applicable. Assessing social sustainability. Though the above discussion provides the general themes and indicators of social sustainability, in supply chains these indicators are highly de- pendent on the nature and context of the business. Since most sustainability research related to 27 transport focuses on environmental aspects, social aspects of sustainability in supply chains are seldom studied (Gold and Seuring, 2011). In response, the social cost perspective includes a great variety of actors not limited to governments and public sectors, but involving each individual in the society. The valuation of external social effects is often made in light of in- vestments in infrastructure and political decisions and involves a kind of monetary estimation of the external social effects caused by transport activities. However, in reality the determina- tion of such costs is difficult; for example, the destruction of a natural habitat due to infra- structural construction cannot be valued in monetary terms. Zhou et al. (2000) describe how social sustainability can be ensured by meeting the needs of the product’s population. This approach holds that fulfilling the demands of the pop- ulation with a certain product promotes social sustainability in continuous process industries. This type of measure can relate to supply chain activities instead of actors, as identified by Hutchins and Sutherland (2008). In the current study, Lammgård (2007) view of defining the demands and needs of freight transportation by investigating the attitudes and preferences of the industry has been applied. This approach indicates the acceptability of sustainable practices involved in supply chains within the context of wood biofuels. The attitudes and preferences of the power plants demonstrate the acceptability of sustainable intermodal activities in wood biofuel supply chains. The reason for focusing on the attitudes and demands of the actors involved in the DH supply chains is that traditional indicators have little importance for these actors. The primary contributions of the DH industry are in terms of environmental sustaina- bility, not social sustainability. It is difficult to pinpoint the labour equity, health care, and educational benefits of actors in wood biofuel supply chains, for the industry caters to the environment instead of issuing social benefits. Nevertheless, safety and employment stability can be highly relevant to actors in wood biofuel supply chains since these companies work in rural areas that have provided jobs and safety to their employees. Another theme that indicates the social sustainability of actors considers their macro social performance in supply chains. 2.4.4. Applying sustainability Barbier (1987) describes how sustainable development involves a trade-off among en- vironmental, economic, and social aspects of a system and claims that it is not possible to optimise all three functions all of the time. Any economic process that depends upon condi- tions such as the even use of resources and services may conflict with the maximisation of productivity and preservation of genetic diversity of the ecological and resource system. Thus, 28 depending upon the objectives, decisions should be made regarding what trade-offs should be given priority, and the process of selecting relevant trade-offs should be adaptable to different scenarios. Individual preferences regarding economic, environmental, and social aspects may change over time and have different degrees of importance in different situations. Variations of economic, environmental, and social goals may also occur depending upon the region, since different regions have different preferences. The dynamic aspects of sustainability based upon time, personal preferences, region, and several other conditions make decisions concerning trade-offs possible based upon objec- tives. In other cases, certain aspects may have more significance than others. Supply chain management poses implications for three aspects of sustainability, depending upon the type of supply chain. For example, an urban supply chain might focus on social and environmental factors rather than economic factors, while a freight distribution supply chain might not be able to induce many social impacts. In fact, social impacts may become less prominent in in- dustrial regions where the majority of business activity occurs among companies focusing on costs. As above-mentioned, making supply chains green may or may not imply cost- effectiveness. However, with advancements in environmental sustainability, significant eco- nomic gains can be achieved by making supply chains green. Reducing packaging and intro- ducing intermodal transportation are just a few examples. 2.5. Sustainable Transport A sustainable transport system (STS) should uphold the three aspects of sustainability discussed at length in the previous section. In general, an STS should increase economic growth and social equity without damaging the environment. The European Commission (2004) has outlined STSs as follows: x An STS allows access and the development of people, organisations, and societies while meeting the safety needs and not deteriorating the health of humans or the ecosystem; x An STS should promote social equity for the current and future generations; x An STS is affordable and functions properly with the chosen mode selection to support a competitive economy while balancing regional development; and x An STS reduces the emissions by restricting the use of non-renewable fuels to the extent at which the planet can absorb them back, as well as minimises emissions and noise im- pacts, ideally by using renewable resources. 29 However, the above definition does not fully align with the concept of sustainability, which holds that non-renewable fuels should never be used in an STS and that the choice of mode should be limited to the best modes and exclude all others (Behrends et al., 2008). 2.6. Intermodal Transport Generally, intermodal transportation involves promoting the benefits of different modes of transportation in a single organised transport chain. Road transportation offers the flexibility and ease of reaching different destinations, while rail and sea transport enable large volumes of goods to be transported over long distances at low costs. A combination of these different modes can help to reduce costs for transporting goods. For example, distribution and collection networks are usually over roads, while longer hauls are normally performed via rail or sea transport. Terminals are used to shift from one mode to another. To make the system efficient, goods are carried in standardised load carriers called intermodal transport units (e.g., containers, swap bodies, semitrailers (Flodén 2007, Bektas and Crainic, 2007). 2.6.1. Definitions The concept of intermodal transport can be narrowed with standardised terms, for the combination of multiple modes has been defined in different terms. The most commonly used terms are multimodal transport, combined transport, intermodal transport, and co-modality (Reis et al., 2013). An early definition of multimodal transport provided by the United Nations (1980) states that international multi mode transport involves utilisation of at least two different modes. In an international construct the goods are taken from one country and delivered to a destination in another country by the use of at least two modes operated by a transport operator. More recently, the United Nations (2001) defined multimodal, intermodal, and com- bined transport. While multimodal transport is ‘the carriage of goods by two or more modes of transport’, intermodal transport is ‘the movement of goods in one and the same loading unit or road vehicle, which uses successively two or more modes of transport without handling the goods themselves in changing modes’. Lastly, combined transport is ‘intermodal transport where the major part of the European journey is by rail, inland waterways, or sea, and any initial and/or final legs carried out by road are as short as possible.’ 2.7. Chain Components Jensen (1990) explains that the key feature of intermodal transport is the combination of cost and service benefits of different modes to improve the overall efficiency of the transport network. Reis et al. (2013) describe intermodal freight transport as the integration of 30 transport agents. Theoretically, the performance of all transport agents can be optimised, though the maximum performance that can be achieved in the real world will always be lower than the theoretical performance. Nozick and Morlok (1997) explain that rail transport in intermodal chains begins and ends at terminals. Goods are transported to terminals by road at both the origin and destina- tion. Goods in intermodal chains are either carried in load units (e.g., containers, trailers). A shipper loads an empty load unit, which is taken to the origin terminal and transported to the destination terminal by rail, accompanied with the necessary routing information. At the des- tination terminal, the train is unloaded and the goods are taken to the consignee by road. Due to uneven flows between the origin and destination terminals, normally positioning the empty and filled load units is required and should be considered in the system. Figure 4 illustrates intermodal chains developed by Flodén (2007). Figure 4: A typical intermodal supply chain. 2.7.1. Terminals Intermodal terminals can be privately or publically owned and include rail yards and air, sea, and river ports. The key roles of intermodal terminals include providing space and equipment to exchange loads between different modes. Other roles of the terminal may in- clude storage and the consolidation and/or sorting of vehicles and goods. Thus, terminals per- form a critical role in the intermodal supply chain, and delays in terminal operations can re- duce the efficiency of the whole chain. Operations at different intermodal terminals (e.g., land terminals, seaports, airports) might differ, though their purposes remain the same. Equipment present at the terminals can differ as well; seaports are often equipped with quay cranes and forklifts, among other equipment (Bektas and Crainic, 2007). Several factors are involved in the success of an intermodal terminal’s operation. The high use of terminals does not necessarily mean that the overall system is efficient. Intermodal terminals require the fast transhipment of goods between different modes, yet have trouble Terminal Long-distance transport Sender / Re- ceiver Collection and dis- tribution 31 with competing for large flows over medium distances, suggesting the number of terminals should be kept to a minimum. For short distances, costs depend upon road transport, which often involves travelling long distances or in opposite directions to reach terminals. Intermod- al terminals thus need to be integrated well into the overall supply chain (Jensen, 1990). The ownership of intermodal terminals is decisive in most transport systems. For- warders often keep the consolidation of goods to themselves, while the movement of goods between terminals is contracted out (Bergqvist et al., 2010). A large flow of goods requires trailers with heavy containers and swap bodies, which in turn require large terminals that are expensive to operate. Such conditions require few large terminals with substantial distance among them for the system to be efficient given high tran- shipment and maintenance costs (Nelldal et al., 2005). 2.7.2. Load units In outlining the various components of intermodal chains, Bektas and Crainic (2007) describe the importance of load units and terminals. Intermodal chains rely heavily on con- tainerisation for its many benefits, including safety and handling. The standard structure of containers used also facilitates the exchange between modes, thereby reducing handling costs. Terminal equipment and operating procedures are often improved to save on costs associated with handling the containers. Some load units used in intermodal chains are discussed in what follows. Lorry and trailer combinations. These units include conventional freight-carrying ve- hicles that tow separate load trailers to form what is known in Europe as a road train. De- pending on weight restrictions, vehicles can have any combination of four-, five-, or six- axle configurations. Trailer axles often include a steerable front with a single or double bogie fitted at the rear. The rear can be separated from the vehicle and stand on its own wheels alone with landing legs. An alternative to such a design includes two or three closely spaced axles, which are non-steerable and located centrally along the length of the trailer. Trailer combinations are generally designed to carry 6 meters ISO containers or a 7.15- or 7.45-m swap body. The skeletal frame of the vehicles includes twist locks for fastening containers or swap bodies (Lowe, 2006). Containers. Freight containers are usually built according to standard dimensions set by the International Standards Organization (ISO), hence the term ISO container. These con- tainers are equipped with twist locks and can be lifted from the top by special container cranes, straddle carriers, or stackers. ISO containers have fork pockets underneath so that they 32 can be carried by heavy-duty forklifts. Standard ISO containers are 6, 9, 12, and 13.7 metres long, 2.4 meters wide, and either 2.6 or 2.9 meters high. They are built to be strong and can usually be stacked eight or nine units high. In Europe, 6 and 12 meters ISO containers are the most commonly used since these measurements fall within the legal dimensions of road transport on the continent. Bigger and/or specialised containers are also used for special pur- poses (Lowe, 2006). Swap bodies. Swap bodies are lifting units with strong material for lifting on and off of road and rail vehicles. Both 7.15- and 7.45-m long swap bodies have fold-down legs that enable them to stand alone. Lifting pockets appear underneath all swap bodies for lifting pur- poses. Swap bodies are not as strong as ISO containers given their lighter construction; how- ever, savings on tare weight and increased payload potential are attractive (Lowe, 2006). 2.7.3. Transportation Every mode of transport can be used in an intermodal transport system. The different modes of transport are road, rail, inland shipping, short sea shipping, deep sea shipping, air, and pipeline. In this study, rail and road are discussed since biofuels are currently transported largely via these modes. Rail freight. Rail transport of large volumes over long distances results in lower costs and emissions. In an intermodal setting, rail transport uses intermodal rail wagons that can carry ISO containers, standard swap bodies, or whole vehicle combinations. ISO containers are usually carried on skeletal flat wagons with low loading heights to carry 2.9 meters tall ISO containers fitted along with twist locks at 6- and 12-meters centres. Whole vehicles are carried on special pocket wagons that keep them securely fastened. The vehicles are kept low so that they can pass through bridges and tunnels (Lowe, 2006). Road haulage. Transporting goods generally starts with a road transport carrying the goods loaded in an ISO container or swap body. The initial choice of vehicle may depend upon the shipper or the road haulier responsible for the movement of goods. Road hauliers may transport goods all of the way themselves or may transfer loads to a rail or ship for the long haul, after which a local road haulier becomes responsible for delivering the goods to the destination. The shipper can arrange such an arrangement as well. Other actors can be in- volved in the intermodal supply chain, including freight forwarders, which do not own any vehicles but contract necessary actors in the chain. These actors act on the behalf of the ship- pers. Freight forwarders exist to make the transport of goods simple and easy for the shipper (Lowe, 2006). 33 2.7.4. Chain characteristics Flodén (2007) elaborates that costs in an intermodal transport system incurred at a terminal must be kept lower than the usual road transport costs. This guideline implies that a practical intermodal transport system should try to minimise the use of terminals and maxim- ise the long-haul transport system. The balance of costs of terminals and long hauls shows that intermodal transport may not be competitive for short distances. A certain distance, which is highly case sensitive between the sender and receiver of goods, is said to be a minimum requirement for any intermodal transport system. However, this requirement can be altered with better management and advances in the technology or equipment, according to different studies. Bergqvist and Flodén (2010) describe how intermodal transport systems have substan- tial potential to reduce CO2 emissions given increased road–rail transport. However, certain challenges such as technical developments, attitudinal changes among actors and the need of differentiated business models still exist. Strong competition and low value transport custom- ers make it difficult to invest in methods to reduce emissions. In this regard, operating princi- ples are needed to develop standardised market requirements. Lowe (2006) summarises some benefits of intermodal transport with single load units: x Low transit costs while covering longer distances; x Fast delivery times; x Reduced traffic congestion; x Less environmental pollution (e.g., noise, emissions, vibrations); and x Reduced energy consumption when long hauls are made via rail or sea. Konings (1996) identifies shortcomings of intermodal transport systems that require attention. For instance, intermodal transport chains often face certain disadvantages compared to other chains, including cost competition with direct road transport in the absence of termi- nal costs. However, this shortcoming is often balanced by the added advantages of handling. For short distances, these costs are higher, which thus limits the minimum distance in the sys- tem, and due to the added costs, it is necessary to save costs in the total cost of the system. Achieving reliability and reduced transit times is also troublesome for intermodal transport, especially at points where loads are exchanged between modes, which to reduce makes stand- ardised load units for different modes seem vital. Furthermore, huge amounts of goods moved via intermodal transport will pressure not only handling terminals but also pre- and post- haulage performed by trucks. A greater number of loads means the increased use of trucks to 34 collect and distribute goods at both ends of the intermodal chain. In response, researchers have sought solutions that promote cost-effectiveness and quality (Konings, 1996). The shortcomings of the intermodal chain are also shared in wood biofuel supply chains as well, in which road transport dominates alongside issues with handling operations. Current policies and the increased demand of biofuels will involve international trade and long-distance transport of goods in large quantities. Transport costs can be determined by vehicle costs, availability, and biofuel characteristics in the case of bulk carriers (Lamersa et al., 2012). Transport is a vital part of biomass supply chains and can substantially add to costs depending upon the geographic locations of the supply and consumption points, as well as transport options (Miao et al., 2012). Though residential areas usually use domestic biofuel resources, much industry growth is anticipated in the power plant sector, which generates both heat and electricity (Hoefnagels et al., 2014). Transport in wood biofuel supply chains con- sumes the most fossil fuels. Studies have shown that using a combination of rail and road can allow the transport of large volumes with less energy use over longer distances (Lindholm and Berg, 2005). 2.8. Swedish Railroad Intermodal Transport System In 1996, the deregulation of rail freight transport in Sweden led to open competition. The Swedish National Rail Administration (Banverket) had owned the railway infrastructure in the country, though such ownership was divided among the railway companies to cultivate competition among railway organisations (Lammgård, 2007). The three major forwarders Schenker, DHL, and DSV and the railway company Green Cargo are principal actors in the Swedish industrial transport system. In the past, the hauling company Bilspedition and a sub- sidiary company of the state-owned railway company called ASG dominated the road haulage sector. Furthermore, Swedish state railway company SJ had a monopoly in rail transport (Flodén, 2007). In the early 2000s, the whole Swedish intermodal network structure was simplified and almost completely dismantled, resulting in all trains operating between just two terminals (Almotairi et al., 2011). The activities of the Swedish Road Administration and Banverket were merged into the Swedish Transport Administration (Trafikverket) on 1 April 2010. The railway network administered by Trafikverket consists of 11,900 km of track, 3,700 km of which consist of multiple or double tracks. The Swedish road network has 98,400 km of state roads; municipal roads consist of 46,500 km, while private roads receiving state subsidies comprise 75,900 km. Around 19,700 km of roads, or 20% of the country’s total road length, 35 are gravel, and approximately 66% of gravel roads appear in Sweden’s forest counties (Trafikverket, 2010a). 2.9. Wood Biofuel Supply Chain Actors 2.9.1. Terminals Terminals used in wood biofuel supply chains can provide different services, includ- ing moisture content measurement, weighing, chipping, and storage. Chipping at a terminal can be performed only if away from residential areas due to the noise and dust produced dur- ing the process (Wolfsmayr and Rauch, 2013). Terminals are specifically used by HPs to ac- commodate seasonal fluctuations and capable of storing chipped and non-chipped wood bio- mass along with by-products from sawmills (Rauch and Gronalt, 2010, Wolfsmayr and Rauch, 2013). The reduction of terminal costs depends upon the entire wood biofuel supply chain, since the involvement of terminals often results in additional costs due to storage and other operations. However, chipping and transport costs can be reduced with high scales of input (Rauch and Gronalt, 2003). Terminals tend to have either permanent or mobile chipping equipment on site. They often have a specified storage capacity for different type of wood biomass. For example, chipped wood biomass must be stored on a hard surface and protected from rain, meaning that different costs are incurred for storing different types of wood biofuels. Another reason is that storage period reduces the energy value of the wood biofuels; as such, their separation needs to be maintained. Terminals near harbours do not provide the possibility of chipping and thus are used only for imported wood biofuels. Often, terminals at plants have limited storage ca- pacity, which requires wood biomass to be chipped upon arrival. Terminals are also said to provide good service in terms of using different modes of transport (Gunnarsson et al., 2004). Required storage time of wood biofuels at terminals depends upon customer demand and contracts. To manage seasonal fluctuations, power plants require terminals to have buffer stocks to bridge the gap between the supply and demand of wood biofuels. A terminal needs to develop substantial stocks before supplying to a new power plant in order to deliver the desired service. This process requires estimating the input stocks that a plant requires to oper- ate, along with the safety stocks that the plant and terminal can secure. With the knowledge of monthly stock needed for a plant, the respective storage capacity and investment can be calcu- lated. Calculated costs should also take into account the maintenance costs of the stocks (Gronalt and Rauch, 2007). Transhipment operations at terminals add fixed and variable costs to wood biofuel supply chains (Mahmudi and Flynn, 2006). 36 Stockyardsterminalen is a prominent example of a terminal that handles wood biofuels in Sweden. The terminal has rail access and consists of 50,000 m2 of storage. The company Stockyardsterminalen AB was founded in February 2008 and is owned equally by the municipality of Sävsjö and Sävsjö Transport AB (Stockarydsterminalen, 2009). Wolfsmayr and Rauch (2013) describe the general type of terminals used in wood bio- fuel supply chains, as follows. Industrial terminals are often located near forest-based industrial plants equipped with a stationary chipper for pulp wood, which can also accommodate forest fuels (Rauch and Gronalt, 2010). Train terminals are often used to store large amounts of wood, which need to be transported for further use over long distances involving multimodal transhipment. Simple terminals are present in forest areas whose sole purpose is to provide storage. 2.9.2. Transport actors Internal competition in truck transport due to its wide use keep transport costs compet- itive. Wood biofuels from forests can be carried to plants in three forms: point-of-origin form, as part of another product, or in reduced particle size. Point-of-origin form involves carrying fuels in their harvested form and is rarely used due to low bulk density and high unit weight transport costs, unless compression techniques are used to increase payloads. This form also increases loading and unloading costs at facilities due to the heterogeneous nature of wood biomass. This form is only suitable for short transport distances between the point of origin and consumption. The form in which wood biofuels are carried along with other products is a commonly used method. Bark on round wood used as wood biofuel is a common example. The tree section method used in Scandinavia is a more specific form that involves leaving branches attached to the trunk. At a pulp mill, both the trunk and branches are taken where branches and bark are separated for energy purposes while the trunk is used for pulp produc- tion. However, if the point of consumption is not present at the pulp mill, transport costs dou- ble. The extra costs derive from the loading and unloading costs of the wood residues at the pulp mill and their transport to the power plant (Hankin et al., 1995). Green Cargo is of special importance in wood biofuel supply chains, as its 2012 com- pany report claims. It satisfies the vast list of Swedish freight transport needs, including fuel chips and pulp woodchips. Green Cargo operations involve rail-based freight transportation complemented by road transport, and the combination of rail and road transport solutions is a key part of its operations. By 2012, Green Cargo had 2,323 employees and provided its ser- 37 vices to various industries such as the steel, chemical, automotive, engineering, forestry, and retail industries, including customers such as Volvo, Stora Enso, IKEA, and ICA. It is owned by the Swedish state and administered by the Ministry of Finance (Green Cargo, 2012). Biofuels in Sweden are increasingly becoming a part of the intermodal transport sys- tem. Some major train terminals that handle wood biofuels are Norrland, Jämtland, Hälsing- land, Medelpad, Ångermanland, Härjedalen, Dalarna, Småland, Halland, and southeastern Norway, as well as in ports such as Hargshamn in Uppland. Biofuels from these terminals are carried to populated areas such as Mälardalen, Nykvarn, Örebro, Eskilstuna, Uppsala, Karls- tad, and Gävle, and potential destinations include Göteborg, Malmö/Lund, Västerås, and Skel- lefteå. Trains are estimated to transport approximately 1,400 GWh of biofuel (including peat) per year, though this is a rough estimate since different units of volume are used. The quantity equals approximately 500–750 full trains, though the size depends upon the type and moisture content of the biofuels. Statistics from Swedish wood fuel producers show that more than 40,000 GWh of wood biofuels were sold in the market in 2011. Green Cargo is the most dom- inating company in the biofuel train transport, followed by Hector Rail; other forwarders in- volved in biofuel transportation are TÅGAB, CFL Cargo Sverige (formerly MidCargo), RushRail, and Inlandståget AB (Hersle and Berglund, 2012). Rotary containers (volume = 46 m3) are primarily used for the train transport of wood biofuels, for which each train carries 60–70 containers (i.e., 2–3 containers on a single car- riage). The containers require a special forklift to be unloaded, and generally, 6-meter con- tainers are rarely used in train transport. Most train cycles comprise 24 h, and the common frequency of train deliveries ranges from one to four trains per week, though some routes have higher frequencies (e.g., 14 trains per week). 38 3. Methodology This section provides an overview of the methodology, along with the data collection tools used for the study. This thesis is inspired by soft system methodology as it moves from the state of identi- fying problems to describing a system, in this case the wood biofuel supply chain. Checkland (1999) states that system thinking is based upon the concept of viewing the world as a system. It is described as: “The existence at certain levels of complexity of properties which are emergent at that level, and which cannot be reduced in explanation to lower levels, is an illustration of an al- ternative paradigm – that of ‘systems’. The systems paradigm is concerned with wholes and their properties.” Identification of the system and problems were followed by the calculation of costs and CO2 emissions in a current wood biofuel supply chain. This helped in identifying the sus- tainable options that can be utilized in a wood biofuel supply chain based upon a case study. The inspiration from soft system methodology helped the study move from different phases of identification of the system to definitions and problems. The study is then concluded with a description of intermodal options that can be applied in wood biofuel supply chains while addressing the various problems in the system. 3.1. Research Theme The wood biofuel supply chain is considered a system involving logistical activities such as harvesting, storage, transport, and pre-treatment activities. The following table pro- vides the research design of the study: Table 2: Research design. Identification and problem description Paper 1 Literature review Paper 2 Description of preferences and constraints in wood biofuel chains Paper 2 Survey Study Improving sustainability in wood supply chains Paper 3 Case Study The first paper describes the wood biofuel literature and thus helped in laying out the problem situation for the wood biofuel supply chains and the logistical challenges found in the literature. The second paper focussed on testing the specific problems, along with a de- 39 scription of the activities involved in a wood biofuel supply chain. The third paper helped in the improvement of sustainable chains for wood biofuel power plants. Figure 5 describes the distribution of the topics in each of the three papers developed. 3.2. Data collection methods This section describes the various data collection tools that were used during the study. The different sets of tools helped in developing diverse results as different tools result in different insights. The data collection tools used are as follows: 3.2.1. Literature review The supply chains of wood biofuels are unique in various aspects and require a thor- ough understanding of the various processes involved in the method of delivery to the energy plants. The literature review carried out as the first part of this thesis helped in understanding not only the concept of wood biofuels but also the wood biofuel supply chain. Blumberg et al. (2005) describe the various purposes of a scientific literature review, which include develop- ing the context of a problem or a topic with reference to previous studies along with several other purposes. In this thesis the literature review had the purpose of establishing the context of the problem and the topic of wood biofuel supply chains. The first paper described the various definitions used for the wood biofuels in the industry and literature. The literature review also provided an overview of the whole supply chain of wood biofuels, along with the already identified problems. The literature review provided the necessary standpoint of understanding the logistics of wood biofuels along with the existing problems. The literature review study utilised the literature, which was present digitally from journal databases along with research reports from other institutions in Sweden. Usage of online resources gave the benefits of sav- ing time in the literature review process and helped refine the search results to find the most relevant material for the topic at hand. The literature review focused on the transportation of wood biofuels on different modes. An effort to provide comprehensive results with a focus on domestic wood biofuel transportation was made. The topic of biofuels includes different types of biofuels, along with issues not only in the field of logistics but also other technical fields. Therefore, defining keywords and selecting specific databases was not an easy task, along with the special focus on the domestic wood biofuel transportation. However, keywords used to search the material studied were “biofuels”, “wood biofuels”, “intermodal transport of biofuels”, “wood biofuel supply chains”, “wood fuels transportation”, “wood biofuels transportation”, “biofuel supply 40 chains”, and “issues in biofuel supply chains”. The databases used were “Science Direct”, “Emerald Insight”, and “Google Scholar”. Since the databases showed different results along with studies related to other geographical locations, a lot of valuable insights were provided by the other research reports from various institutions and organisations, such as the Swedish University of Agricultural Sciences, The Swedish Energy Agency, and Statistics Sweden. 3.2.2. Survey study A survey is a great tool for obtaining primary data and learning opinions and attitudes (Blumberg et al., 2005). Some researchers consider designing a questionnaire more of an art rather than a science. Such researchers believe that there is no best way to design a question; however, different phrasing and formats provide different results, and all of these results are equally valuable in understanding the respondents (Krosnick, 1999). A web-based survey was distributed via e-mail during the summer of 2013 to a com- plete sample of managers at all 76 existing biofuel-using CHPs in Sweden. Selection of CHPs was made due to the fact that they utilize large quantities of biofuels. Such characteristics make them ideal for a study of intermodal transport of wood biofuels. Online surveys have several advantages, as Blumberg et al. (2005) points out, which includes ease in the develop- ment and analysis phase. Online surveys are also cost efficient in comparison to other meth- ods, such as surveys delivered via post, and it allows for branched surveys more adapted to the respondent’s situation. In addition to the various benefits of web-based surveys, there are some drawbacks associated as well. The most prominent would be the ease with which the respondents can decide not to answer the survey with just the click of the button. This draw- back is common to all survey studies; however, web-based surveys are easier to get rid of in comparison to other methods of data collection. A pre testing was performed for the survey with a few respondents, which provided vi- tal insights to improving the various questions in the survey. The survey was divided into two parts. The first included questions about the equipment and practices at the power plants. These questions covered the general topics of storage, transportation, chipping, and overall supply chain design. The second part investigated the perception of the power plants towards current and desired practices regarding the transport of wood biomass, which involved rank- ing the different practices or issues on a scale. The survey is provided in appendix. The survey comprised of closed questions with the option of open comments. Scale questions were used in order to determine the attitudes of the respondents towards different supply chain activities, issues, and practices. 41 Approaching the total population helped perform a census study. Blumberg et al. (2005) describe a census study as: x “feasible when population is small, x necessary when the elements are quite different from each other.” A total of seventy-six CHPs exist in Sweden; they have a small population and are easily approachable, especially via email. The second condition for a census study - for the respondents to be quite different - is not quite possible for CHPs, as they have common char- acteristics, such as storage terminals. Minor differences between the plants can be observed, such as some plants utilize storage terminals away from the plants while others use on-site storage. Similarly, sourcing wood biofuels is different for different plants, as some receive their stocks from the forests while others receive by-products from sawmills. However, due to the small population it is ideal to contact the whole population to get comprehensive results. Such conditions make a census study a suitable approach for studying the supply chains of wood biofuels. In the current study, respondents who were reluctant to answer were approached through telephone and urged to answer. The total response rate was 42% (n=32). The survey contained 30-38 questions and was made adaptive to the answers given. For example, if the respondent did not use a certain type of fuel the survey adapted and removed related ques- tions. 3.2.3. Case Study Yin (2009) describes case studies as rich in empirical descriptions of particular in- stances of a phenomenon that are typically based upon different data sources. Case studies are excellent for building empirical evidence, which helps in building understanding of a particu- lar phenomenon and seeking its application to a more general level. Case studies provide the simplicity and ease of trying and analysing different scenarios before implementing them on a general level. For this purpose a case study was developed based upon the operations at a plant in Gothenburg, Sweden, owned by Göteborg Energi. The plant provides an opportunity to study intermodal and sustainable chains as it has both rail and road access, along with storage ca- pacities vital in the wood biofuel supply chain. The case study investigates the various costs and environmental effects of transporting wood biofuels to the plant based upon different quantities and sources of wood biofuels. This would be achieved by analysing different sce- narios in terms of costs and CO2 emissions for supplying the plant with wood biofuels. 42 Social sustainability has been left out of the case study as wood biofuel supply chains have few details on what can be considered socially sustainable. Labuschagne et al. (2005) described social sustainability as involving three major themes: internal human resources, external population, and macro social performance. It is difficult to define the social perfor- mance of the supply chains of CHPs except in terms of providing employment in rural areas. The CHPs’ contribution to the social sustainability is highly limited as the plants are rarely involved in welfare activities. Defining social sustainable aspects of wood biofuel supply chains are also difficult as this involves investigating the attitude of the whole Swedish popu- lation towards the use of wood biofuels, which would require a separate study. The contribu- tions of the wood biofuel supply chain actors to social sustainability would also require de- tailed study with different actors focusing on the main themes of social sustainability. The methodology for the case study involved designing a potential rail delivery sys- tem and then subjecting it to a sensitivity analysis in which different variables were changed in order to find the key factors influencing the intermodal transport. The potential rail system is designed by finding a break-even point between the road and rail transport. In this regard a search for the local terminals and sourcing options was made in order to have a detailed rail system. The variables were later changed to see the influence on the system, which helped in the formulation of a best-case scenario. Storage levels were subjected to a similar analysis in which scenarios of missing train deliveries were assessed. In this scenario the after effects of missing train deliveries were analysed and various options available were also discussed. The detailed cost and emission data can be found in appendix. 3.3. Validity and reliability The research process consists of the measurement tools required to obtain the neces- sary data. It is vital for the measurement tools to be not only accurate but effective as well. Reliability and validity are two important aspects of any research. Detailed attention to these concepts marks the differences between a good and a poor research, thus developing the cred- ibility of the scientific results. In a qualitative research perspective these sections need extra attention as the scientific society requires detailed explanation of the concepts in perspective to the study (Brink, 1993). Validity and reliability are not necessarily symmetrical, as it is possible to obtain reliable results without any validity; for example, a broken thermometer may give results but they will not be valid. Conversely, perfect validity yields perfect reliabil- ity, as absolute truth is measured. In social science reliability is assured by various means; however, since perfect validity is not even theoretically possible this always leaves room for 43 improvement (Kirk and Miller, 1986). Validity and reliability are described as follows and their role has been discussed in the light of the current thesis. 3.3.1. Validity Concept of validity relates to the truthfulness and accuracy of the findings in the re- search process (Lecompte and Goetz, 1982). A valid study gives results, which exist in reali- ty, and the instruments used in such a study measures exactly what they were designed for. Many different forms of validity are mentioned in the literature; however, the two basic types of validity are external validity and internal validity, as described by Campbell and Stanley (1963). Denzin (1970) applied the distinction between the two types of validity to qualitative studies. The truthfulness of a study and the extent to which the study reflects the reality is referred to as internal validity. External validity refers the extent of the results applicable across different groups (Brink, 1993). In the current study internal validity has been emphasized, as compared to external va- lidity. For the current study external validity would require validation of the study’s findings from additional studies. Applicability of results to different groups requires further studies, which can be carried out in the future. Internal validity, however, is described in detail and has been ensured to the most possible extent in the current study. Internal validity can be further divided into three different groups: content, criterion, and construct. Content validity refers to the selection of the right content to be asked of the respondents. Analysing the right content is highly judgmental and can be approached in many ways, all while requiring clear understanding of the topic under study. A panel of experts can also be used to judge the relevance of the content used in an instrument to collect data. Nunnally (1959) described criterion validity as the ability of a measurement procedure to generate results similar to an alternative procedure, which is valid. Blumberg et al. (2005) refer to criterion validity as an instrument’s ability to predict or estimate. In order to have cri- terion validity the instruments should be relevant, free from bias, reliable, and available. Rel- evance aspect of a measurement tool requires it to cover the necessary topics related to the study. Freedom from bias refers to the aspect that respondents can give uninfluenced infor- mation and their analysis should be done on equal terms. Finally, the availability aspect refers to the availability of information required by the measurement tool. Cronbach and Meehl (1955) describe criterion validity as the ability of a measurement tool to generate results, which can correspond to the theoretical aspects of the phenomenon studied. Blumberg et al. (2005) elaborates that in order for a measurement tool to have con- 44 struct validity it should be in line with the theoretical definitions of the phenomenon. For ex- ample, in order to measure ceremony in an organizational culture, the tool should be aligned with the definition of ceremony in the theory. The three different types of validity are interrelated; however, which type of validity needs to be focussed on depends upon the nature of the study. It would be hard to say that any of the types of the validity can be ignored completely since they are interrelated. In the current study the measurement tools used include the three types of validity in certain ways. A literature review study was conducted to have the required topics needed to be covered in the survey and the case study in order to have content validity. In addition to the literature review, the survey study also went through a panel of experts that provided their insights on the topics to be included or omitted from the survey. Defining key search words helped in obtaining the relevant articles, which insured the relevance aspect of criterion validity in the literature review. Freedom of bias of the articles was maintained by selecting articles from different journals, thus reducing a certain focus of the journal in the phenomenon under research. Since all of the articles and reports used to gather the relevant information are published and available online, the reliability and availa- bility aspect of the criterion validity is also fulfilled. Surveys maintained a relevance aspect of criterion validity by not only taking ad- vantage of the information gathered through the literature review but also by going through a panel of experts that contributed and assured that the content of the surveys would generate information relevant to the supply chains of wood biofuels. Freedom from bias of the surveys was assured by the use of adaptive online surveys. The online adaptive surveys selected the relevant questions depending upon the characteristics of a plant. For example, if a respondent did not have access to rail transportation, no rail related questions were asked so that specula- tion could not be part of the information gathered. During the development phase of the sur- vey special consideration was given to the aspect of speculation so that such questions would not be asked of a respondent who might not know the answer to a specific question. Plant managers were the target respondents of the surveys, which maintained the information avail- ability aspect of the criterion validity of the topics covered in the surveys. Criterion validity in the case study followed the measures taken during the develop- ment of the survey study. The relevance aspect can be seen by the characteristics of the plant selected for the case study, such as access to intermodal transport options along with high consumption of wood biofuels. Freedom from bias can be a critical part of a case study, as the 45 company under investigation can influence the content covered in the research. Therefore, separate academic and company report documents were developed in order to keep the aca- demic and industrial focus separate during the case study. The calculations and analysis done in the case study are based upon a previously developed model and was academically valid. Usage of the previously published and known model reinforces the reliability of the results generated in the case study. Criterion validity of the study lacked the predictability of the phenomenon measured. The instruments were designed to give a description of the phenome- non under study rather than the prediction of the future. In the literature review focus on the articles relevant to the subject area of logistics and wood biofuels ensured construct validity. The survey study maintained construct validity by focusing on the current operations of power plants and their attitudes towards various sustain- able activities. This helped in attaining the objectives of the survey study set by the research questions. The results of the case study highly relied on the selection of the already published and established model, which helped in getting accurate results. 3.3.2. Reliability A general definition of reliability is the consistency of a measurement procedure to give results, which correspond to the results of a similar procedure. Comprehensive under- standing of the reliability can only be achieved by specifying the research process completely from which scientific results have been generated (Meyer, 2010). Abell et al. (2009) argue that a measurement tool is reliable if it produces similar results used independently among different groups. Reliability means different things to different studies, depending upon the nature of the study. In general it is said to be the consistency of the results obtained from a study. Reliability is an important aspect to the results but may not insure the validity of the results. Consider a bathroom scale that measures one weight 3 kg more every time it is used. The number given by the scale is reliable but it is not valid as it gives the incorrect weight. Measurement tools need to be reliable in order to give accurate results with changes in time and conditions. The dependence of reliability over time and conditions highlights the three aspects of stability, equivalence, and internal consistency in a measurement tool. An instru- ment is said to be stable if it can obtain the same results from the same person over and over again. Equivalence refers to the consistency of the measurement procedures for a similar group of respondents. Equivalence depends upon the measurement tools being used and the way in which they are used. Internal consistency describes the homogeneity in the items in- cluded in the measurement tools. The correlation between the items used in a measurement 46 tool can be said to be representative of the internal consistency. Homogeneity of the items used in a measurement tool also depends upon their relevance to a certain phenomenon under study (Blumberg et al., 2005). The measurement tools used in the current study involved surveys and interviews. Since the tools have not been assessed over a time period it is hard to define their reliability in terms of passage of time. However, the results obtained from the two tools are consistent and complimentary for the different topics discussed by each tool. The use of web surveys also insured the equivalence aspect of reliability by providing the same surveys to the different respondents, thus removing any chance of error in conduction of surveys. Interviews, in con- trast, followed a list of topics identified through the use of the literature study and included topics that were discussed in the survey as well. Since the current study is descriptive in na- ture it is not possible to describe its reliability in terms of prediction. 47 4. Summary of appended papers The following section provides the results obtained from the literature review, survey study and case study. In addition, a section 4.8 is included that is based on the work in a re- port that is not appended to the thesis. 4.1. The appended papers in brief Three papers are appended to the thesis. In addition, the research has been presented in a number of project reports. 4.1.1. Wood biofuels logistical challenges in Sweden Author: Awais, Fawad. The basic purpose of the paper is to review the current literature on wood biofuels and identify key logistical challenges for different types of biofuel supply chains. The paper in- corporates an investigation of the definitions used for wood biofuels and their raw materials in research on the logistics of wood biofuels. The generally-used distribution networks are iden- tified and analysed. The paper concludes with a consideration of the various logistical chal- lenges that are present in the wood biofuel industry. Information and data are obtained through a literature review study. This paper contributes to the field by analysing the supply chain for wood biofuel from a holistic perspective and providing a solid foundation for further research on how these challenges can be solved. Findings: This paper introduces definitions related to wood biofuels and raw materials. It also outlines the various logistical problems that arise in the distribution of wood biofuels and raw materials for use in heating and power plants in Sweden. Key challenges identified include seasonal variations, storage, the chipping process, the low density of wood biofuels, term standardisation, supply sources and dependence on policies. 4.1.2. Logistic requirements and characteristics of the Swedish wood biofuel industry Authors: Awais, Fawad & Flodén, Jonas. Sweden has a substantial utilisation of forest fuels in district heating plants and com- bined heat and power plants. This places large demands on the logistics system supplying these plants with fuel. The aim of this paper is to identify the effect of industry actors’ re- quirements, constraints and preferences on the wood biofuel supply chain and to identify the logistical challenges this entails. To achieve this, a survey was sent to all Swedish CHPs, and 48 six interviews were conducted with transport companies, terminal operators and forest com- panies. Findings: This study shows that the industry has a local-market focus, mostly utilising truck transport with direct transport from the forest. Road transport is rated highly favourable, with reliability as the most important factor. Storage is used to overcome fluctuations in de- mand and is an essential part of the supply chain, and most CHPs have storage facilities. The forest is the preferred location for chipping. Challenges include determining the size and loca- tion of storage facilities and identifying transport alternatives that might improve the transport chain and reduce environmental impact, while at the same time maintaining flexibility. 4.1.3. Meeting the challenges for intermodal transportation of biofuel Author: Flodén, Jonas & Awais, Fawad. The use of solid biofuels for energy in heating plants has increased drastically in re- cent decades. This substantial and increasing demand has drawn focus to delivering the sup- ply to the plants, as logistics issues are considered one of the key challenges for further in- creased use of biofuels. Environmental concerns, the increasing size of power plants, and challenges in sourcing enough fuel locally have sparked an interest in using intermodal road- rail transport. A case study was conducted at a Swedish district heating plant to investigate the potential for introducing intermodal transport. Extensive calculations were performed for the design and operation of an intermodal system, showing both costs and CO2 emissions. The results are analysed in relation to key logistical challenges in the industry, and a best feasible case scenario is identified. Findings: Conclusions are that the potential for intermodal transport is greatest among the largest plants with large volumes to achieve high resource utilisation. An advantage of intermodal transport is that large flows currently pass through a terminal, which improves the competitiveness with road transport and allows for the use of efficient resources at the termi- nal. This study leads to a better understanding of the strengths and weaknesses of intermodal biofuel transport and has practical implications for anyone in the process of designing such systems. 4.1.4. Project reports This thesis is part of the project titled, “Sustainable Intermodal Supply Systems for Biofuel and Bulk Freight,” which involves research about the intermodal transportation of biofuels. Therefore, separate reports were generated in collaboration with partners in the pro- ject. The reports most helpful to the compilation of this thesis are as follows: 49 x WP 1 Report: Biofuels, Authors: Fredrik Bärthel, Fawad Awais and Jonas Flodén (Editor) from School of Business, Economics and Law, University of Gothenburg, Dag Hersle and Moa Berglund from WSP. Aim: The aim of this report is to introduce the concept of bio-based materials. Biofu- els and the characteristics of these goods are mapped. This mapping is followed by a literature review incorporating past research, market information and current intermodal transport sys- tems and supply chains for bio-based materials in Sweden. A concluding analysis is made. x WP 2 Report: Logistic Requirements and Characteristics in the Swedish Wood Biofuel Industry. Authors: Fawad Awais and Jonas Flodén from School of Business, Economics and Law, University of Gothenburg. Aim: The aim of this paper is to identify the effects of industry actors’ requirements, constraints and preferences on the wood biofuel supply chain and to identify the logistical challenges this entails. x WP 3 Report: Designing Intermodal Supply Chains Authors: Jonas Flodén (Editor), Johan Woxenius, Fawad Awais and Jon Williamsson from School of Business, Economics and Law, University of Gothenburg, Moa Berglund, Helena Billing Clason and Dag Hersle from WSP,Johanna Enström from Skogforsk, Behzad Kordnejad from Royal Institute of Technology, KTH. Aim: The aim of this report is to analyse the logistical challenges in designing a sus- tainable intermodal supply chain for biofuel. The concept of sustainability is explained and the characteristics of each transport mode for biofuel transport are evaluated. Business models in the industry are mapped and intermodal business models are further developed. The possi- bility of return flows is investigated, followed by a case study of a large power plant. x Report to Göteborg Energi: Possibilities for intermodal transport of biofuel in Sweden, a case study on Göteborg’s Energi plant Sävenäs. Authors: Jonas Flodén and Fawad Awais from School of Business, Economics and Law, University of Gothenburg. Aim: The aim of this report is to investigate the potential use of the intermodal transport of biofuel for use in HPs. This study leads to a better understanding of the strengths and weaknesses of intermodal biofuel transport and has practical implications for anyone in 50 the process of designing such systems. The study is set under Swedish conditions as the use of biofuel for HPs is particularly well-established in Sweden. These reports were a vital part of shaping this thesis and therefore must be mentioned for the sake of better understanding of the concepts discussed in the current study. 4.2. Logistics in wood biofuel transportation Gold and Seuring (2011) describe two main objectives of the biomass supply chain: keeping biofuel costs competitive and ensuring continuous supply. The different logistical steps in the wood biofuel supply chain can be divided into four steps. These four steps, along with descriptions, have been summarized in Table 3. Table 3: Various steps of SCM/logistics and bio-energy. Adapted from (Gold and Seuring, 2011). Category Description Collecting/ harvesting This part of the supply chain is located in the forest where the wood is collected, and is more focussed on forest management than logistics. The chipping process taking place in the forest affects the handling and cost of the wood biofuel transportation. Pre- treatment activities This step involves the chipping, drying and conversion of wood biomass to densified biofuels such as wood pellets. These activities have economic, environmental and social impacts on the other operations of the biofuel supply chain. Transport Transport takes place between various points of the wood biofuel supply chain, such as the forests, storage terminals, heating plants, etc. The form of wood biofuels along with the mode used for transportation are key components. Storage This is a key step in the supply chain as it is useful when encountering seasonal variations in the demand and supply of wood biofuels, as well as providing another point for the chipping process. Overall supply system This refers to the challenging task of effectively and efficiently designing and operating bio-energy production. Most of the literature on biomass supply focuses on the overall supply chain rather than components. Energy plants and biomass suppliers are the most important actors in the chains and are interdependent. The logistics of the biomass supply chain are more complex than for fossil fuels, as the biomass supply chain involves more deliveries (Gold and Seuring, 2011). Consider one TWh, which is equivalent to 86,000 tonnes of the equivalent of oil. One tonne of wood pellets holds 4.6 megawatt hour (MWh) of energy, which means one TWh corresponds to 217,000 tonnes of wood pellets (Andersson, 2012). Low transport costs help the biofuel plants to work at optimum levels. Economies of scale and increased efficiency of logistical activities are critical to the biomass supply chain (Gold and Seuring, 2011). Gunnarsson et al. (2004) describe the different decisions involved in wood biofuel chains, such as which fuel type to use, the time required for the movement of the biofuels, 51 whether or not to employ the process of chipping, the location of the chipping process, storage at terminals, the design of the transportation network, the need to establish contracts with the forest and sawmill owners and restrictions regarding the capacities for chipping, forwarding and storage at terminals. Gunnarsson et al. (2004) also identify key strategic planning situa- tions for the supplying companies involved in the wood biofuel chains: submitting competi- tive prices, finding a solution to varying demand for wood biofuels—e.g. in colder-than-usual months, adopting a new storage terminal or altering the capacity of the already existing stor- age terminal, altering the capacities of the chipping sites and the modes of transportation, changing costs due to the change in the chipping technology and negotiating with transport companies. A general representation of the transportation activities involved in a wood biofuel supply chain is presented in Figure 5, which was developed by the author for this study. Figure 5: National supply chains under focus. Triangles represent onsite chipping and dotted lines represent chipped biomass. 1* Operation 1: Harvesting and collecting biomass 2* Operation 2: Storage 3* Operation 3: Transport 4* Operation 4: Pre-treatment techniques 52 The arrows in the diagram represent the flow of the wood biofuel raw materials and products. The triangles in the diagram represent the onsite chipping process, e.g. the triangles adjacent to the forests represent roadside chipping done in the forest. The logistical activities that take place between the origins and destinations of the wood biofuels include transport and intermediary activities such storage. It should be noted that transports may be carried directly from the location of origin to the power plants or may pass through storage terminals. 4.2.1. Harvesting and collecting biomass Harvesting and collecting biomass is the first step in the logistics of any biofuel supply chain. Transportation and chipping play a vital role in the production of wood biomass. Utilis- ing efficient equipment and methods can lead to reduced costs—for instance, the use of chip- per trucks can reduce overall costs by 20-40%. These chipper trucks can be competitive for distances of 100-120 km (Björheden et al., 2010). Routa et al. (2012) note chipping location as a major factor in the overall layout of the wood biofuel supply chain. 4.2.2. Storage Storage is required to ensure a smooth supply of biofuels. The number of storage ter- minals generally increases if the harvest period is short (Gold and Seuring, 2011). Stockpiles along roadsides can be considered a type of storage. Storage plays a vital role in the drying of wood biofuels without extra costs. Unprocessed wood biomass can be left outside, whereas processed wood biomass such as wood pellets needs to be stored in covered places (Hamelinck et al., 2005). Wood biofuel demands are the highest during autumn and winter; therefore, the biofuels are stored for at least few months. The moisture content normally de- creases with increased storage time, although it may also rise due to weather conditions, which may mean even longer storage times (Anerud and Jirjis, 2011). 4.2.3. Transport in the bio-energy chain Transport issues related to the bioenergy chain can be divided into categories of infra- structure framework, transport operations and social and environmental impacts (Gold and Seuring, 2011). Road transportation is utilised for short distances or as the initial or final haulage of biofuels over longer distances for plants that lack rail and sea infrastructure. Using trucks to satisfy the demand of large CHPs can result in congestion problems (Mahmudi and Flynn, 2006). Johansson et al. (2006) describe the typical road vehicles used in Sweden as having a maximum length and width of 25.25 and 2.60 metres respectively, with an allowable carrying weight of 60 tonnes. Longer distances and huge volumes can balance out the costs incurred due to rail and sea transportation for large power plants. The factors that influence 53 the costs for the rail transportation of wood biofuels are deliveries made by the train units per unit time, the utilisation of load capacity, transport distance, storage terminal handling (load- ing and unloading) and conditions of train shunting at storage terminal points (Björheden et al., 2010). Two common container systems for wood biofuels are described in Table 4. Table 4: Main loading and unloading techniques of wood biofuels (Björheden et al., 2010). System Advantages Disadvantages Rotary system Specially designed containers mean greater volumes on each wagon. Efficient unloading with turntable forks. The unloading forklift truck is specialised and has few other uses. The forklifts are designed for rotary con- tainers and cannot handle other containers. Container sys- tem for switch- body trucks Long-distance transportation of containers by trucks is possible, as containers are designed for both trucks and trains. Includes standard components. Longer loading and unloading times are required compared to the rotary system. Load capacity cannot be utilised to the maximum of the train, as the weight and width limit of the containers on trucks and trains are different (trains have wider lim- its). 4.2.4. Pre-treatment techniques Pre-treatment activities for wood biomass include drying and pelletisation. Drying helps reduce the moisture content of the wood biomass, thus improving the combustion effi- ciency along with transportation. Pelletisation involves converting biomass to densified mate- rial, improving energy content and handling. Open-air drying does not incur any costs, how- ever it is dependent upon climatic conditions. Pelletisation incurs extra costs in the process of converting wood biomass to pellets. (Gold and Seuring, 2011). 4.3. Logistical Challenges Rentizelas et al. (2009) highlight that a significant amount of the costs of biofuel ener- gy are due to the logistical activities involved in the process. Storage is a major concern, mainly due to the seasonal variations of the biofuels. The different forms of biomass require different equipment for handling (e.g. chips require different handling equipment than logs). The list of logistical challenges discussed below was selected based on the literature review, and the challenges are described according to the literature. 4.3.1. Seasonal variations Carlsson and Rönnqvist (2004) note that seasonal variations impact the amount of stock present in the wood biofuel supply chain. In Sweden, the traditional harvesting period 54 for wood biofuels is autumn, winter and early spring. The difficulty of harvesting wood dur- ing winters along with growing energy demands result in large stockpiles during spring and very small ones during summer. The period between August and early September is critical, with the possibility of a shortage of wood biomass before the harvesting process begins again. Gold and Seuring (2011) elaborate that seasonal variations in the demand and supply of wood biofuels result in the underutilisation of equipment and labour. Furthermore, the shorter the harvest periods, the greater the period of stock retention, resulting in higher storage costs and dry matter losses. The large volumes of stock in winter are suited to rail transportation, but rail transportation cannot be used in summer when the demand is low. 4.3.2. Storage The necessity of storage facilities for the wood biofuel chains involves a number of logistical decisions regarding the location, size and equipment present at a storage facility. Storage facilities deserve special attention as they directly affect the overall cost of the supply chain. 4.3.3. Chipping process Hamelinck et al. (2005) argue that the chipping process strongly affects the transport and handling costs of the wood biomass and presents decision-making scenarios. Chipping is typically performed at the roadside, the terminal or the power plants. The physical form of the wood biomass affects handling and transportation activities and costs, so where chipping oc- curs has implications. Asikainen (1998) notes that chipping at a terminal or plant is only effi- cient for large volumes due to the high fixed costs of the large chipping equipment, whereas when chipping at the forest the equipment can be hired on an hourly basis. Terminal and plant chipping provide more control over the quality of the wood chips and require no setup costs or allocated space for chipping equipment, while in forests the chipping location can vary due to weather and the surrounding conditions. Waiting times due to delays may result in added costs and reduced efficiency. 4.3.4. Low density of wood biofuels Wood biomass and the wood chips produced from it are low-density goods and require capacity management in vehicles and at storage terminals, resulting in different handling. As illustrated in Figure 6, different forms of the wood biomass occupy different amounts of space but contain the same amount of biomass. 55 Figure 6: The differences between different types of biomass (Björheden et al., 2010). All loads con- tain the same amount of biomass, but in different forms they occupy different amounts of space. The low energy density of the wood biofuels makes transportation an important cost factor in the supply chains (Gold and Seuring, 2011). Bradley et al. (2009a) note the problem of the low density of wood biofuels in ships. 4.3.5. Term standardisation Frosch and Thorén (2010) argue that the lack of standard terms of wood biofuels makes it difficult to estimate the consumption and production of wood biofuels within differ- ent regions. A standardised catalogue of the various wood biofuels would help enable clear comparisons and trade among the wood biofuel entities. Jungingera et al. (2008) argue that a lack of standardised terms in the direct trade of round wood also generates complications, as the sawdust produced from round wood is also used for energy generation. 4.3.6. Sources of supply Hirsmark (2002) points out that increases in the demand of wood biofuels in exporting countries can affect the imports of wood biofuels to Sweden. Imports can also be affected by increases in international demand. If more countries demand wood biofuels with no more suppliers, the prices of wood biofuels will rise in local markets, and biofuel quantities will shrink. 4.3.7. Dependence on policies Jungingera et al. (2008) point out the dependence of biofuel utilisation on policies. Eu- ropean Union member states are trying to increase the use of biofuels for the purpose of achieving sustainability. However, some cases show that varying policies can result in in- creased biofuel prices while disturbing the market mechanism. Selkimaki et al. (2010) also 56 highlight that Swedish authorities have shaped their policies for the propagation of biofuels in the district-heating sector, mainly due to fossil fuel taxation and subsidies. 4.3.8. Logistical challenges identified from survey study The challenges mentioned in the literature were also investigated in detail in the sur- vey study to test their validly, and as a result the challenges were narrowed down to the major ones. The survey also helped by ranking the challenges from most to least problematic. The following tables provide the ranking of logistical challenges identified in the survey study. Table 5: Issues in the Swedish biofuel industry (1=least problematic, 6=most problematic). Issue Mean rating Seasonal variations in the demand for heating 4.3 Dependence on Swedish political decisions on biofuels 4.0 Dependence on foreign policy decisions on biofuels 4.0 Biodegradation of the biofuel 3.9 Contamination of the biofuel at delivery, e.g. stones 3.8 Drying of the biofuel 3.3 Impact on the Swedish biofuel market from increased global demand for biofu- els 3.1 Cooperation between actors in the industry 3.0 Standardised terms and definitions for biofuels and raw materials 2.7 Availability of biomass 2.6 The location and size of the storage facilities are significant issues regarding the stor- age of wood biofuels. The following ranking of storage issues is for storage facilities both at the plant and at off-site storage facilities. Table 6 provides the ranking of storage issues pro- vided by survey respondents. Table 6: Mean ranking of storage problems (1=least problematic, 6=most problematic). Storage issues Total Small CHP Medium CPH Large CHP Size of storage facilities 3.8 3.8 3.5 4.8 Location of storage facilities 3.7 3.4 3.5 4.8 Handling of biofuel at storage facil- ities 3.4 3.2 3.3 4.0 Availability of equipment at storage facilities 3.2 2.9 3.4 3.5 57 Table 7: Mean ranking of transport problems (1=least problematic, 6=most problematic). Total Small CHP Medium CHP Large CHP Low density of biofuel 3.6 3.3 3.7 4.0 Risk of contamination of the biofuel dur- ing transport, e.g. stones 3.5 3.6 3.3 3.3 Transport of tree parts 3.1 2.9 3.4 3.5 Transport of forest residues 3.0 2.8 3.3 3.5 Availability of suitable transport 3.0 2.7 4.0 2.3 Transport of wood chips 2.5 2.5 2.7 2.5 Transport of pellets 2.3 2.1 2.5 3.0 The ranking of transport issues in Table 7 shows that the transport of wood biofuels is considered more problematic by the larger CHPs with larger flows. Tree parts and forest resi- dues are the most problematic fuels. The ranking corresponds well to the density of the fuels, itself ranked as a top problem. The survey revealed that on a more general level, the largest problem perceived was the seasonal variation in demand (see Table 5). In general the market appears well-functioning with, no major cooperation problems or lack of fuel. 4.4. Swedish wood biofuel logistics This section describes the results of the survey with a description of the industry situa- tion in Sweden. The total energy produced by the CHPs in the survey was 12.6 TWh (9.4 TWh in winter [October-March], 3.2 TWh in summer [April-September]), or 26% of the total energy production (48.1 TWh) by CHPs and HPs in Sweden (Svensk Fjärrvärme, 2010). The majority (74%) of this type of energy is produced during the winter. The respondents can be classified into three sizes based on winter energy production as shown in Table 8. Table 8: CHP size. Table 9 shows the percentage of energy produced by different fuels according to the survey distributed among the Swedish combined heating plants. Fuel use is dominated by logging residue chips followed by other wood chips. Wood residue chips were the most wide- ly used fuel, with 75% of the respondents using this fuel, followed by chips of other wood (68%) and stem ships (50%). Essentially the same fuel mix is used in summer and winter. Plant size in winter Percentage of population Number of respondents Small (0-250 GWh) 50% 16 Medium (251-500 GWh) 31% 10 Large (>500 GWh) 13% 4 Unknown 6% 2 Total 100% 32 58 Table 9: Fuel used. 4.4.1. Operation 1: Harvesting and collection Plants know the origin of the wood biomass they receive, as prior planning is done to secure the supply. Contracts are generally established on a yearly basis, often with the same suppliers. Local suppliers and buyers are preferred, thus creating monopoly/oligopoly rela- tionships. The preference for local sourcing is understandable, as it lowers transport costs and ensures quick deliveries, however, larger plants have to source over longer distances. Pur- chase of fuel is commonly made in terms of energy content, though some traditional units persist (e.g., firewood is by tradition traded in solid m3). Competition from other industries is expected to present risks due to economic fluctuations or increased production of bioethanol (as it requires raw materials such as sawdust and popular tree species). Exports are not con- sidered feasible at the moment due to the high prices of biofuel in Sweden along with a lim- ited demand for biofuels in the rest of Europe. This limits the import of biofuels as well, but import is economically possible despite the challenge of keeping transport costs low. 4.4.2. Operation 2: Storage Storage was an essential component for most of the respondents, and it is present both on the plant premises and at other locations. Tables 10 and 11 highlight the storage locations and time of the power plants. Storage is typically found at both the plant location and at an off-site location. The average storage time in winter is low because of the high demand as compared to the low demand of summer. Fuels Heat generated in winters Heat generated in summers Stem chips 9% 6% Logging residue chips 24% 22% Stub chips 2% 1% Chips of other wood 21% 25% Chips of unknown wood 9% 10% Pellets 2% 3% Peat 7% 3% Waste burnt to produce energy 8% 16% Bio gas 0% 0% Natural gas 0% 0% Bio oil 1% 0% Coal 1% 0% Oil 5% 0% Other fuels: 11% 14% Total 100% 100% 59 Table 10: Share of respondents having storage. CHP sizes At the CHP At the CHP and nearby Nearby No storage All 59% 28% 6% 6% Small 69% 25% 0% 6% Medium 80% 20% 0% 0% Large 0% 75% 25% 0% Table 11: Average storage time in days. Storage location Average storage time (days) in winter Average storage time (days) in summer CHP storage All 36 55 Small 50 63 Medium 26 49 Large 3 60 Close-by stor- age All 89 137 Small 113 114 Medium 137 183 Large 47 131 4.4.3. Operation 3: Transport Swedish forests and wood industries (sawmills, paper mills, etc.) produce the majority of the wood biofuels used at power plants. The supply chain design is dominated by direct shipments from forests to power plants. Use of terminals results in extra costs and is avoided if possible. Table 12 shows the percentage of respondents using various supply chain designs. The table also elaborates the energy that is produced by using different supply chain designs, helping to illustrate the feasible supply chain structures currently in use. 60 Table 12: Transport chains used. Winter Summer CHPs using the chain Energy produced CHPs using the chain Energy produced Directly from the forest 69% 26% 63% 39% From the forest via storage terminal 34% 12% 19% 3% From the forest via transhipment terminal 21% 7% 15% 8% From the forest via both storage and tran- shipment terminal 7% 6% 7% 6% Directly from a forest industry, such as sawmills 59% 23% 56% 21% From a forest industry via storage termi- nal 3% 0% 4% 0% From a forest industry via transhipment terminal 3% 0% 4% 0% From a forest industry via both storage and transhipment terminal 3% 6% 4% 6% From abroad 14% 1% 7% 1% Unknown chain 24% 19% 22% 16% Table 13 shows that road transport is the most commonly used mode of transport; larger plants have access to other modes of transport such as rail and sea, but these are rarely used. The daily number of trucks for large plants can range between 50-70 trucks per day. Road transport distances are generally less than 250 km and are kept minimal due to the low value and energy density of wood biofuel. The management and presence of unloading and loading equipment at both terminals and plants are not an issue for the supply chain. 61 Table 13: Transport distances. T ru ck o nl y T ra in o nl y Sh ip o nl y T ra in , t he n de liv er y by ro ad T ru ck , t he n de liv er y by tr ai n T ru ck , t he n sh ip , t he n de liv er ed b y tr uc k D el iv er y by tr uc k, p re vi - ou s st ep s un kn ow n D is ta nc e (k m ) T ot al < 10 0 N = 23 10 0- 25 0 N = 15 25 0- 50 0 N = 3 > 50 0 N = 3 U nk no w n N = 3 T ot al U nk no w n T ot al 50 0- 75 0 T ot al N = 25 0- 50 0 N = 1 50 0- 75 0 N = 1 T ot al U nk no w n T ot al N = < 25 0 N = 1 U nk no w n T ot al < 25 0 N = 1 Percent of respondents 94 75 50 9 6 22 3 3 6 3 3 3 3 3 3 9 3 6 6 3 Percent of energy in winter 84 47 17 3 3 14 2 2 1 1 4 2 2 5 5 2 1 1 2 2 The percentage of respondents in Table 13 represents the percentage of a certain mode or combination of modes from the total population. Therefore, the sum is not at 100%, as many respondents use different modes or combination of modes. This means that a respond- ent may be using both a “road only” option and a “rail only” option. The results show that a certain mode or combination of modes is represented by a percentage of the total population. As is often the case in logistics (e.g.,Lammgård (2007); Saxin et al., (2004)), the most im- portant factor for the respondents is reliability. Respondents were asked to rank different fac- tors associated with transportation, and environmental sustainability received the lowest rank- ing in terms of importance. However, most CHPs are municipality-owned and are thus re- quired to follow certain purchasing and environmental standards. 62 Table 14: Ranking of important modal choice factors and service received (1=least ful- filled/important, 6=most fulfilled/important). Service Importance Service received Total Small CHP Medium CPH Large CHP Total Small CHP Medium CPH Large CHP Low cost 4.7 4.5 4.9 5.0 3.8 4.1 3.5 4.0 Short transport time 3.8 4.0 3.4 3.8 4.5 4.6 4.4 4.8 High reliability 5.1 5.1 4.9 5.5 4.7 4.9 4.3 4.5 High frequency 4.2 3.9 4.0 5.0 4.6 4.9 4.4 4.5 Few contaminations in the fuel, e.g. stones 4.9 4.8 4.5 6.0 4.1 4.4 3.6 4.0 Environmental sus- tainability 4.5 4.0 4.7 5.8 3.9 4.3 3.2 4.0 Good access to the transport system, such as infrastruc- ture 4.7 4.3 4.9 5.8 4.7 4.7 4.6 4.8 Table 14 describes the percentage of respondents that agree with the characteristics of different transport modes. The results have been further broken down based on the categorisa- tion of plant sizes. The percentages shown are the respondents in agreement with a certain characteristic of a particular mode. Road transportation was rated as having the maximum benefits, though it was rated poorly in terms of environmental sustainability and transport costs. Rail and sea transport, on the other hand, lack the benefits of road transport but are per- ceived as environmentally sustainable. Large power plants have more experience with rail and sea transport, as can be seen in their ranking of the various benefits of these transport modes. The perception of the various characteristics presented in Table 15 helps with understanding the important factors in considering the selection of transport modes. The preference of modal choice factors has been mostly paired with road transportation, as shown in Table 15. 63 Table 15: Qualities of different transport chains. Qualities Total Small CHP Medium CHP Large CHP Trucks have low cost 16% 13% 10% 50% Trucks have short transport time 52% 50% 60% 50% Trucks have high reliability 64% 60% 70% 100% Trucks have high frequency 55% 50% 60% 75% Trucks have good access to the transport system, such as infrastructure 68% 63% 80% 75% Trains are environmentally sustainable 48% 40% 60% 75% Trains have none of the specified char- acteristics 42% 50% 40% 0% Ships have none of the specified char- acteristics 45% 50% 40% 25% A combination of trucks, trains and ships have none of the specified char- acteristics 61% 63% 60% 25% A combination of trucks and trains have none of the specified characteris- tics 58% 63% 40% 25% A combination of trucks and ships have none of the specified characteris- tics 52% 13% 10% 50% The ranking of preferred transport modes in Table 16 shows a very clear advantage for road transport and supports the preference for truck transport. Table 16: Mean ranking of preferred transport mode (1=least preferred, 6=most preferred). Mode of transport Ranking Road 5.5 Rail 2.5 Truck and ship combined 2.5 Ship 2.3 Truck and rail combined 2.3 Truck, rail and ship combined 2.2 Of respondents, 37% and 41% consider the use of rail and sea transport, respectively; however, the majority of CHPs let their suppliers arrange the transport modes to be used for delivery. Table 17 shows that the most commonly used load units for the delivery of wood biofuels are fixed/tilting trucks followed by switch-body trucks. The most common load unit for rail transport is the rotary container. These statistics are helpful for the determination of preferred load units by power plants. 64 Table 17: Load units/vehicles used. Load units for trucks Percentage of energy produced Percentage of usage Small CHP Medium CHP Large CHP Truck with fixed/tilting superstructure, e.g. side-dump truck 67% 75% 100% 100% Container truck with switch body, e.g. wood chip container 25% 69% 90% 100% Other 6% 19% 0% 25% Do not know 2% 6% 0% 0% Delivery times for the biofuels are flexible along with unloading times for the different modes. Trucks have the lowest average time for unloading, followed by trains. 4.4.4. Operation 4: Pre-treatment techniques The chipping of wood biofuels is mostly performed in the forest, and this is also the option preferred by most respondents. Chipping is often considered a part of harvesting (Gold and Seuring, 2011), but can also occur at other locations in the chain as a pre-treatment activi- ty. Table 18 reveals that forests are the preferred location for chipping, though large plants prefer to chip at the plants in order to have better control over the quality of the wood biofu- els. Local restrictions can affect chipping on plant premises, as power plants may be located in residential areas. Table 18: Share of energy produced by various chipping locations and their preference (1=least pre- ferred, 6=most preferred). 4.4.5. Overall operation of the supply chain Uneven demand and supply is a major challenge for wood biofuel supply chains, and Table 19 shows what options plants use to handle this challenge. The fluctuating demand for biofuels is handled similarly by the different categories of plants. Cooperation with other Total Small CHP Medium CHP Large CHP Chipping location C H P s us in g S ha re o f en er gy P re fe re nc e C H P s us in g P re fe re nc e C H P s us in g P re fe re nc e C H P s us in g P re fe re nc e Chipped at the forest 66% 42% 4.6 53% 4.4 80% 5.1 75% 3.8 Chipped at the terminal 55% 18% 4.1 47% 4.6 60% 3.8 50% 3.0 Chipped at the CHP 41% 19% 3.4 27% 2.8 50% 3.6 75% 5.0 Chipping location unknown 41% 19% - 53% - 20% - 25% - 65 plants and storage of produced energy are more common among the larger plants, while smaller plants rely more on buffer stocks and deliveries. Table 19: Handling fluctuations in demand, with percentage of respondents using the option. Table 20 shows what transport services are important to the power plants and to what extent these services are currently provided. The most important factors in the biofuel supply chain are on-time deliveries, fuel quality and no contamination in the fuel. Comparing the preferences and services received reveals that the CHPs are in general satisfied with their supply chain. Table 20: Services received and their importance (1=least fulfilled/important, 6=most ful- filled/important) Service Service received Importance Total Small CHP Medium CPH Large CHP Total Small CHP Medium CPH Large CHP Flexibility concerning delivery options 4.3 4.3 4.3 4.3 4.3 4.3 4.1 5.0 Flexibility concerning ordered volumes 4.4 4.8 4.1 4.3 5.0 4.9 5.1 5.3 On-time deliveries 4.4 4.8 4.4 3.8 5.4 5.4 5.2 5.8 Low transport cost 3.6 3.9 3.4 3.3 4.8 4.6 4.9 5.0 Low biofuel price 3.5 3.9 3.2 2.8 5.4 5.4 5.7 5.0 Small environmental impact 4.2 4.3 4.1 4.3 5.0 4.8 5.2 5.5 Good quality biofuel 4.5 4.5 4.5 4.3 5.5 5.6 5.4 5.8 No contamination in the delivered fuel, e.g., no stones 4.1 4.1 4.1 3.5 5.6 5.6 6.0 6.0 Deliveries evenly distrib- uted over time 3.9 4.0 4.0 3.0 4.8 4.7 5.0 5.0 4.5. Case of Sävenäs Power plant The plant selected for the case study is located at Sävenäs in the city of Gothenburg, Sweden, and is owned by Göteborg Energi. The plant provides an ideal situation to study in- termodal transport chains as it has access to rail transport and has a relatively large consump- tion of wood biofuels. In addition, Göteborg Energi provided full cooperation and funding in Options used Total Buffer stocks 74% Extra deliveries 71% Energy storage 35% Use of fossil fuels 29% Cooperation with other heating plants 13% 66 order to study their currently used and potential supply chains. The plant was constructed as a coal plant to supply heat in 1985, and was later converted to a biofuel plant. 4.5.1. Case introduction The Sävenäs plant has access to a rail network, although a full-length train has to be split, as the rail sidings are just 200 meters long. The plant is located adjacent to a residential area, which has certain restrictions that affect the plant. Chipping is forbidden by local envi- ronmental regulations. Fuel deliveries can be made to the plant during weekdays and week- ends, but weekend deliveries are avoided to prevent complaints from the neighbouring resi- dential areas. The plant receives deliveries between 6 A.M. and 10 P.M., resulting in a 16- hour window for deliveries during weekdays. The plant is currently fuelled by wood residue chips, log chips and stump chips. The plant is currently supplied entirely by road through local sourcing. About forty trucks are used daily to deliver wood biofuels to the plant. Figure 7: Local track layout and adjacent shunting yard. The plant produces 17 GWh of energy every week (seven days) when working at full capacity. The plant has on-site storage that is 10,000 m3 in volume and holds fuel for about 60 hours of energy production. In our assumed scenario, the plant can alternate between two types of train deliveries. One train contains wood chips and the total energy carried by this train is 2.1 GWh. The train coming the next day carries 1.75 GWh of energy in the form of bark wood. These foundational facts are helpful in determining the various possibilities for road and rail operations. Storage at the plant is 10,000 m3, which in the form of grot chips or log- Unloading area Plant 67 ging chips is equivalent to 7.79 GWh of energy with minor variations. The 7.79 GWh of stor- age can provide 3.2 days (7.79/2.429) of energy production. Table 21 summarises some basic facts about the power plant under study. Table 21: Energy when plant operating at maximum capacity. Total energy produced in a week 17 GWh Total energy produced daily 2.43 GWh Total storage in terms of energy (in form of chips) 7.79 GWh Number of days that can be supplied by storage 3.2 days Energy carried by a wood chip train 2.1 GWh Energy carried by a bark train 1.75 GWh Energy carried by a truck 0.093 GWh The assumption that a train can carry 2-2.5 GWh of energy implies that 150-190 trains are needed annually, or roughly one train per day during the operational season. Like all sys- tems, train deliveries are vulnerable to unexpected events such as delays, accidents or shut- downs. Therefore, total dependence on train deliveries is not preferable. The non-electrified local rail siding is also a challenge for the plant, as it requires time-consuming and expensive shunting operations. Rail transport is not currently utilised at the plant—a successful rail transport test was conducted, but the associated costs were very high. The test included standard all-round 35m3 containers. The containers were transferred to a switch-body truck by forklift. The unloading operation by truck was efficient, but the con- tainers did not utilise the full capacity of the train. The shunting and splitting operations re- quired for the train to enter the short local rail sidings were also time-consuming and expen- sive. 4.5.2. Case methodology Once the information on current plant operations along with current challenges was laid out, it was time to move to the analysis phase of the case study. The next step involved the estimation of the costs and CO2 emissions during the different processes in the plant’s supply chain. Microsoft Excel was used to develop a modelling tool based upon Flodén (2011) for the case study. Considerable attention was given to the input data for the model, and particularly to the cost data. Six telephone interviews and an email interview were con- ducted to better understand the operations. The interviews conducted were with road (1), rail 68 (1) and sea (e-mail) biofuel transport companies, and with a terminal company (1), a forest company (1) and energy companies (2). The interviews were 60 minutes long and were rec- orded. Input data were collected from the literature (Table 22) and from the Swedish biofuel industry actors. Original cost data were obtained from four industry actors regarding different parts of the supply chain. Variations in cost estimates were expected between different data sources. Combining the data from the literature and from the industry refined variations in the cost data. The resulting data were checked against the author’s own calculations of ex- pected costs. The resulting data set contained a reasonable representation of the costs to be included in the case study. This data can be viewed as an average cost level in the Swedish industry rather than any actor-specific data. The selected data were validated by at least two independent industry representatives, and some data were validated by as many as five indus- try representatives. Selected cost data were also validated by a reference group of biofuel in- dustry actors from road, rail, power plant and forest sectors. Cost estimations were made in SEK, Swedish kronor, kr (2014: approx. 9 kr = 1€). Table 22 Cost literature sources Data Literature sources Assumptions Biofuel densities Larsson and Nylinder (2014), COFORD (2003). Swedish conditions. Road costs Skogforsk (2011), Flodén (2011) Waiting times, etc. considered. Empty returns. Rail costs Flodén (2011) Waiting times, etc. considered. Emissions from Swedish electricity mix. Terminals Asmoarp (2013), Skogforsk (2011), Sommar (2010), Bäckström et al. (2009) Chipping Johansson and Mortazavi (2011), Eliasson et al. (2012), Lombardini et al. (2013) Bioenergiportalen (2013) 69 4.5.3. Break-even distance A break-even distance, above which the intermodal transport was less costly than all- road transport, had to be developed in order to demonstrate the economical viability of inter- modal transport. The plant is currently supplied by trucks only, which meant a certain scenar- io has to be developed in which intermodal transport could be said to be economically ac- ceptable. The break-even distance also depends on the equipment used. In this regard, the typical biofuel train was considered to consist of 22 Sgns wagons with an engine type Rd and rotary load units carrying 2.3 GWh of logging residue chips. A pre-haul of 50 km by road with a capacity of 0.093 GWh was also considered in the rail system of wood chips. This pre- haul distance was considered because the forest or the sourcing locations seldom have railway lines and have to be transported a short distance to be loaded on a train. Diesel shunting is considered at the terminals and heating plants. Three train deliveries are considered per week for the 26 active weeks per year. The empty return costs were also included. Roadside chip- ping is considered in every case. A break-even distance of 250 km was calculated through analysis involving cost comparison of the all-road solution and three train deliveries per week. The break-even distance represents the minimum distance at which the intermodal distance becomes competitive with the all-road solution. Figure 8 shows the break-even comparison of intermodal transport and the all-road solution. Figure 8: Break-even analysis of road and intermodal solutions. The break-even distance becomes 180 km if five train deliveries per week are made. CO2 emissions are less in intermodal chains compared to the all-road option. Trucks and the chipping process generate the majority of CO2 emissions in an intermodal chain. 70 4.5.4. Base scenario The determined break-even distance of 180-250 km resulted in the selection of the two nearest sourcing areas of Småland and Dalarna, shown in Figure 9. The selection of these sourcing locations was made due to their minimum distance above the break-even distance for intermodal transport. Five train deliveries per week consisting of three train deliveries from Småland (265km) carrying wood residues and two from Dalarna (471) carrying bark was con- sidered for the base scenario. The 180-250 km break-even point provides a safety margin, and is a realistic approach as it provides the option of being economically feasible with the least number of trains per week. Certain unexpected scenarios such as strikes or weather conditions can hinder the de- livery process. Therefore, two sourcing locations were considered in order to have a smooth supply of wood biofuels through the use of rail transport. The selection of the two sources demonstrates that few suppliers can provide alone the amount of fuel needed to make train transport viable. Småland can provide logging residues, the most commonly used biofuel in Sweden, while Dalarna, with a large number of wood industries, can provide the second-most commonly used biofuel, wood by-products. The system can provide 58% of the weekly de- mand of the power plant when operating at full capacity, i.e. 9.8 GWh/week. Göteborg Energi has validated the developed base scenario. Figure 9: The plant and sourcing locations. The star denotes the power plant. Dalarna Småland Gothenburg 71 This base system cost consists of all the processes involved, such as chipping, road haulage, terminal operations, rail costs, etc. The base scenario includes road haulage of 40 km to the terminal at the sources with trucks carrying 0.093 GWh of energy. Chipping takes place in the forests for wood residues, and bark does not require chipping. Terminal operations at Småland involve electric shunting, while Dalarna uses diesel shunting. Fuel at both sources is loaded with a wheel loader, which takes four hours to load a full train. The train used in the base scenario consists of 20 Sgns wagons pulled by a type Rd electric engine with sixty 45-m3-load units with rotary unloading. A train carrying logging residues with these characteristics would carry 2.1 GWh of energy, while a comparable train carrying bark would have an energy value of 1.75 GWh. The train arriving at the plant utilises diesel shunting and takes four hours to be unloaded by a rotator. The base scenario cost calculated for the potential intermodal system is 99 kr/MWh. The rail costs are 35 kr/MWh, which involves shunting, rail transport and load units. The CO2 emissions of the system were calculated as 2.92 kg/MWh, with the majority of CO2 generated by road transport and the chipping process. Figure 10 shows the cost allocations of the chain in which high costs are associated with chipping and the sending terminal. Figure 10: Costs and emissions in the base scenario. 4.6. Base scenario variations Different variations in the base scenario have been considered in the different parts of the supply chain. The variations are detailed in the appended paper three along with the sys- tem costs based upon the variations. 72 The variations in the supply chain activities are based upon the possible options in which each activity can be carried out. The variations helped in the calculations of different costs and CO2 emission estimates and helped in the selection of a best feasible case scenario. The graphic representation of the rail and system costs is summarised in Figures 11 and 12. 73 Figure 11: Summary of costs based on variations in the base scenario. 74 Figure 12: Change in costs and emissions from the base scenario for tested cases without outliers. 75 4.7. Best feasible case scenario The cost calculations done based on the variations can be used to find the best feasible case scenario for the plant, which has high utilisation along with low system costs and emis- sions. A seven-day train operation would deliver 14 GWh or 82% of the plant’s needed fuel at full capacity. However, the plant would rarely operate at full capacity, and thus a five-day per week operation is more realistic, with three days to Småland and two days to Dalarna. The seven-day train operation would also pose risks to trains as it would be hard to manage a 24- hour cycle with tight schedules and with little room to deal with uncertain situations as strikes, bad weather, etc. Chipping of round wood is highly efficient in comparison with log- ging residue chips regarding Småland and has been selected for the best feasible scenario. Large trains with new engines are not selected as they are large in size which are not suitable for the current tracks and infrastructure at the plant, however they would lower costs by increasing the loading capacity and volume delivered (15.2 GWh or 89% of maximum need) by trains. The extra volume provided through the use of new engines is not needed, as they will bring stocks greater than the need of the plant. Older engines and wagons may re- duce costs, but they are more susceptible to failure and should not be included in the best fea- sible case scenario. Thus the engine used in the base scenario, the type Rd electric engine, has been selected for the best feasible case scenario. Rotary containers are considered to be the most effective load units. Return flows are hard to find and thus cannot be included in the best feasible case scenario. Reducing road haulage distances would reduce costs, but the distance of 40 km in the base scenario can be seen as the realistic option as it is the average distance to terminals today. Selecting a supplier with rail access would reduce costs therefore, in the best feasible case scenario bark is trans- ported from Dalarna from a supplier having rail access. The use of trains outside the system and a longer season would further reduce costs. In this best feasible case scenario, a slightly longer season of 28 weeks (vs. 26) is assumed. The rest of the aspects are kept the same as in the base scenario. The distribution of costs and CO2 emissions are graphically presented in Figure 13. 76 Figure 13: Distribution of costs and CO2 emissions for the best feasible case scenario. The best feasible case scenario gives a cost of 77.89 kr/MWh with CO2 emissions of 1.79 kg/MWh. A rough break-even point against all-road transport is at 106 km. Train costs are 36.13 kr/MWh. As major CO2 emissions are generated by the chipping process and road haulage in the chain, electric chipping rather than diesel-powered chipping would greatly reduce emissions, though electric chippers require large volumes to stay economical. If these large use volumes could be maintained, a switch to electric chipping would also reduce costs. In the current sce- nario, the emissions could be drastically reduced to 1.18 CO2 kg/MWh and costs to 68.20 kr/MWh with electric chipping at the terminal. The same principle can be applicable to elec- tric shunting compared to diesel-powered shunting. In addition to electric shunting if rail sid- ings can be extended to avoid splitting the train that will further reduce operational costs. 4.8. Supply risk analysis The supply risk analysis is introduced in addition to the appended articles as a part of the report submitted to Göteborg Energi. The analysis elaborates on the risks associated with missed train deliveries and builds on the effects on the storage and consumption of the plant if certain numbers of train deliveries are missed in a week. The intermodal system delivers 9.8 GWh per week or 58% of the plant’s demand when operating at full capacity. From a risk perspective, the plant is therefore not completely dependent on the train. Figure 14 shows the storage levels at the end of the day when all deliveries have been received. As the plant con- sumes fuel all day, there can be a risk of low stored levels until the deliveries have been made, as shown by the theoretical minimum in Figure 15, with deliveries being made at the closing time. Obviously that is not possible, however it indicates the risk of late deliveries. Figure 16 77 shows the stock level on a Monday, assuming that the train arrives in the late afternoon. Road deliveries are evenly spread from Monday to Friday. Consumption is assumed through the week and across the day. It is assumed that the storage is completely filled by Friday evening, which is the normal procedure since the plant tries to avoid deliveries during the weekend. Figure 14: Deliveries and storage levels (evening) in the base scenario. Figure 15: Example of possible storage and deliveries during one day. Due to seasonal variation, the plant would not operate at full capacity all the time. The minimum point at which the plant is considered to operate is 30% of maximum capacity, be- low which the plant needs to be shut down. Figure 16 demonstrates the increase in costs with half-full trains over a number of weeks. The reduction in demand can be tackled by reducing the fuels brought in or by cancel- ling train deliveries. This is difficult as contracts are made in advance with these possible con- 78 siderations. Whatever strategy is adopted, it will result in an increase of transport costs, as cancelling a train would mean bearing costs due to contract terms with the operator. A sensitivity analysis has been presented based on the number of trains not operating at full capacity. The train costs remain the same as full trains apart from a reduction in energy consumption. Terminal, handling, pre-haulage and other costs are reduced due to lower vol- umes. System costs increase by 36% from 99.95 kr/MWh to 135.90 kr/MWh when the train is 50% full, while rail costs increase by 87% from 35.18 kr/MWh to 65.84 kr/MWh Figure 16: Increasing costs based on number of weeks with 50% full trains. The different storage scenarios are based on the absence of one or more train deliver- ies in a week, which can happen due to strikes, technical breakdowns or inclement climate conditions. The different scenarios developed based upon the absence of train deliveries are as follows: 1. One, two or three consecutive train deliveries missed. 2. Four or five consecutive train deliveries missed. The storage is considered to be 7.79 GWh on Friday, which is the maximum stock (desirable weekend stock), and this serves as the starting point for the scenarios. The daily delivery for one train needs to be complemented with 15 trucks, which are sourced locally to provide the remaining energy for the level considered desirable. During the weekends, the stock levels are reduced to 2.933 GWh of energy as no deliveries are made. The same pattern of one train and 15 trucks can be ordered the next week as well. Through such a combination, by the end of the second week a stock of 2.933 GWh is achieved, the same as the first week, and thus the combination become repetitive and sustainable. 79 4.8.1. One, two or three consecutive train deliveries missed In the case of one missing train not much needs to be done, as the new stock level of 5.69 GWh at Friday can be maintained over a number of weeks without consequence. Missing two trains does not present a major problem, as the stocks only become negative on Sunday only if no extra deliveries are ordered. Thus, in the case of one or two missing trains, compen- sation can be accomplished by ordering extra trucks or by train deliveries over the weekend. The number of trucks can be raised to 21 (in case of one missing train) or 29 (in case of two missing trains) for the remaining weekdays. This will bring the stock levels to 7.79 GWh on the Friday of the first week, which is a normal stock for weekend operations. Missing three consecutive trains will cause stocks to be depleted by Wednesday, and the plant would fail to provide 0.033 GWh of energy to be produced at full capacity. In this scenario, the absence/presence of weekend train deliveries would result in different options. The deficiency could be filled with extra trucks. If the option of weekend train deliveries is not possible, then the 15 trucks that were ordered normally would need to be raised to 37 trucks daily for the last three days of the week. This would bring the stock levels to the nor- mal 7.79 GWh at the end of the first week. If weekend train deliveries are possible, then extra trucks should be ordered only to deliver the fuel needed for production until Friday. It is of no use to have the maximum number of deliveries by truck as trains on the weekend can bring more stock and are a more sustainable option than maximising truck deliveries. 4.8.2. Four or five consecutive trains missed If four trains are missing for the first four consecutive days of the week, then the stock will run out by Wednesday. This means that either stock should be ordered on Wednesday or the plant should have ordered more trucks after missing the first train delivery of bark, i.e. more trucks should have been ordered on Tuesday. The situation would be grim if the plant did not order extra trucks on Tuesday or Wednesday. Options available in this case are either to order the entire stock by truck without weekend deliveries, or to order limited stock so that weekend train deliveries could replenish the stock. The deficiency would be filled with the extra number of trucks. The 15 trucks that were ordered previously would need to be raised to 28 trucks daily for the first week. In the case of five trains missing, the number of trucks daily would need to be raised to 33. This would bring the stock levels to 7.79 GWh at the end of the first week. The follow- ing week the usual 15 trucks and one train per day could be ordered. This results in the same ending stock of 0.0029 GWh on Monday, which can be recovered as stated earlier. If week- 80 end deliveries are considered for trains and trucks, then a different combination is needed. It is not possible to have truck deliveries throughout the week, as if the 2.2 GWh is not ordered on Wednesday then the stocks would be negative, and the objective is to bring the stock levels on Monday to the normal of 4.059 GWh. 4.8.3. Missing train analysis at the Sävenäs plant The number of train deliveries missed is more important than the day of week they are missed for one week. Since five train deliveries are made in a week with large volumes, miss- ing stocks for at least a couple of days can be compensated for by the remaining deliveries. The missing deliveries can also be compensated for by extra truck deliveries, which can be distributed evenly throughout a week. Thus the frequency of train deliveries is more important than the specific day on which a delivery is missed. Extra trucks can handle the missing train deliveries, or weekend deliveries can make up the lost loads; however, these solutions are hard to implement on short notice, especially for rail transport. The option of ordering extra trucks depends on the availability of trucks. In this case, weekend deliveries can be used, as the trucks might not have any engagements during the weekends, as the usual schedule will not include weekend deliveries, but this may not always be the case. The plant can also lease trucks to fetch goods from the terminals. This approach would definitely result in more costs to the plant and would result in under-utilisation of the vehicles when train deliveries are being made smoothly. The waiting period for the plant to decide upon extra deliveries is very limited, as seen in Table 23, as stocks come quite close to deple- tion if no deliveries are made—thus prior planning is required rather than on-going manage- ment. Table 23: Number of trucks and trains needed in different scenarios. One train missing Two trains missing Three trains missing Four trains missing Five trains missing Latest de- livery point No specific delivery point Saturday evening Tuesday evening or Wednesday morning Tuesday evening or Wednesday morning Tuesday even- ing or Wednesday morning Number of trucks per week 100 118 141 160 182 Extra deliveries would reduce both environment sustainability and economic sustaina- bility Replacing trains with trucks would result in more CO2 emissions and costs along with possible congestion problems. The absence of trains could lead to the need for up to 183 81 trucks per week. The option of having daily train deliveries may not be always possible in real situations, but theoretically this can be attained due to a 24-hour cycle. The train delivery timings would become important in this case, as at least an 18-hour gap is needed for another train delivery to be made at the plant from the moment a train is completely unloaded. The unloading times of the vehicles at the plant are not a major concern, as the unloading times for trains (four hours) and trucks (15 minutes) are within the working hours of the plant as long as the vehicles arrive significantly before closing time. An assump- tion in this case needs to be that trucks and trains should arrive and get unloaded at the plant before the closing time (22:00). The minimum time period in this regard would depend upon the number of vehicles required by the plant on a particular day. The time at which the vehicle arrives at the plant is also important for the lead times of the vehicles. For example, if a train arrives at the plant in the evening (no later than 18:00, as four hours are needed to unload the train), it would not be possible to have the same train back at the plant before the 24 hour wait time based upon the sourcing locations mentioned in the case study. The train timetable will affect the return of the same train depending upon the locations from which biofuels are picked up and delivered to the plant. This means the same train can only arrive back after 18:00 the next day. A different train could be used to deliver biofuels to the plant before that time, however that would depend upon train timetables and availability. The train schedules, the unloading times of the vehicles and the arrival times of the vehicles all affect the storage levels. It would be favourable for sustainability to replace the missing trains with extra trains, as this would result in lower emissions and costs, but it may not be a realistic option as the train schedules are mostly pre-defined and it is not possible to order extra trains on short notice. Having extra trains available in the system to just replace missing train deliveries would mean significant costs and under-utilisation of resources. The number of trucks depends upon the availability of trucks, as during peak season (winter) it would be hard to get extra trucks because of their busy timetable. 82 5. Conclusions and future research The following section provides the conclusions to the findings mentioned in the previ- ous section about the appended papers. 5.1. Research questions answered The different research questions developed were answered through the development of the three articles. The summary of the findings is presented in Table 24, along with the rele- vant research questions. The first article identified the key definitions, activities and issues in the wood biofuel supply chain in the literature. The second article helped by describing the industry situation in terms of applied practices and opinions. The third article suggested sus- tainable solutions based on cost calculations and CO2 emissions calculations for supplying wood biofuels to a large power plant in Gothenburg. Table 24: Summary of the research questions along with their answers developed through different papers: Research questions Paper Short Answer RQ 1: What are the different actors and practices involved in wood biofuel supply systems for heating plants? Paper 1: Wood biofu- els logistical challenges in Sweden The supply chain of wood biofuels consists of five steps, which are collection/harvesting, storage, transport, pre-treatment activi- ties and overall supply chain design. The main actors are the for- est sector, the wood processing industry and district heating. The supply chain faces a number of problems of which key challeng- es are seasonal variations, storage, the chipping process, the low density of wood biofuels, term standardisation, sources of supply and dependency on policies. RQ 2: What are the main preferences, requirements, and logistical challenges in the wood biofuel supply system for heating plants? Paper 2: Lo- gistic re- quirements and charac- teristics of the Swedish wood biofuel industry The survey study shows that the industry has a local market fo- cus, mostly utilising truck transport with direct transport from the forest. Road transport is rated as highly favourable, with reliabil- ity as the most important factor. Storage is used to overcome fluctuations in demand and is an essential part of the supply chain with most CHPs having storage facilities. The forest is the preferred location for chipping. The challenges to the supply chain are determining the size and location of storage facilities. On a more general level, the largest problem perceived was seasonal variation in demand. Transport of tree parts and forest residues is the most problematic. RQ 3: How can sus- tainable intermodal transport options be designed for a wood biofuel supply sys- tem for heating plants? Paper 3: Meeting the challenges of intermodal transportation of biofuel High utilisation and keeping transport distances short play an important role in keeping costs down. Large emissions are asso- ciated with the terminals and the chipping process. Some activi- ties in all road systems and intermodal systems are common, such as the passage of goods via terminals, which affords an in- termodal system potential.. Storage levels play an important role in the ordering of shipments. 83 5.2. Logistics processes A clear understanding of biofuel market operations would aid managers in planning and managing the wood biofuel supply chain (Roos et al., 2000). The actor’s preferences and supply chain characteristics, as described in the previous section, can be used to highlight important challenges for the logistics system. 5.2.1. Operation 1: Harvesting and collection The local supply options are clearly visible in the supply chains. Trucks are suitable for shorter distances while other modes require higher transport volumes. This restricts the market for smaller CHPs who source from a distance of 100-150 km, while larger plants have more sourcing options. Large CHPs do prefer local sourcing as well, if possible. Lack of local sources is the main reason for sourcing from longer distances. There are clear differences in terms of procurement between products directly from the forest and the forest industry. The chains starting in the forest are pull chains, in which consumers order what they need. On the other hand, the products from the forest industry have a push chain, where the industry production determines the pace of fuel production. Contracts normally state that plants will use all the by-products from the forest industry, as they are waste for the forest industry. The power in the forest-based chain lies with the CHPs, while in the forest industry chain, the industry has the most power as it sets the pace of fuel production. 5.2.2. Operation 2: Storage Logistically, storage helps in the efficient and smooth flow of fuels and in the handling of excess fuel, which may not be needed immediately (e.g. deliveries from ships and trains that may bring fuels in excess of the daily consumption). Although storage provides flexibility in the supply chain, it is also a major cost factor. Storage demands significant space as well, which is not always possible for the CHPs, resulting in offsite storage options. Rentizelas et al. (2009) conclude that cheap storage solutions can help in reducing overall costs, which could even compensate for the material losses and increased handling at the storage terminals. 5.2.3. Operation 3: Transport A combination of rail and road can transport large volumes of biofuel with lower en- ergy use over longer distances (Lindholm and Berg, 2005). The CHPs in the survey prefer reliable transport with good access, low costs and few contaminations in the fuel, and per- ceive that they receive this from road transport. Using road transport to satisfy the demand of large CHPs can result in congestion problems (Mahmudi and Flynn, 2006). The survey study 84 showed that the large plants could receive up to 70 trucks per day. Shifting the loads from roads to other transport modes is one alternative solution. The survey demonstrated that rail and sea are perceived as environmentally sustaina- ble, but have problems in price, access, flexibility and other areas in comparison to road transport, which is rated low in terms of environmental friendliness. Ships designed for ore transport have handling problems with full loads of wood pellets, and must carry other heavy materials as well (Bradley et al., 2009a). Application of intermodal solutions require large volumes and transport distances, as rail/sea transport has low distance-dependant costs but high fixed costs (Flodén, 2007). Long-distance intermodal transport solutions are slower than the local road solutions, but survey respondents rated fast transport to be of little importance. Suppliers usually decide the transport mode to be used, thus making them an important actor in the supply chain, however abrupt changes in the chain would require approval from the CHPs. The variable demand of biofuels presents transport problems in terms of under- utilisation of machinery causing unnecessary costs. The transport preference of the CHPs found in the survey corresponds well with other industries, as transport quality and reliability were rated as highly important. Transport selec- tion is a two-step process in which first the preferences have to be met, after which the transport options are evaluated on price (Flodén et al., 2010). 5.2.4. Operation 4: Pre-treatment techniques Terminal or plant chipping is high in productivity and requires large volumes to stay competitive in comparison with forest chipping. This produces good quality control with no setup time or allocated space needed for chipping, but delays dues to waiting times can reduce efficiency, resulting in extra costs (Asikainen, 1998). The chipping location greatly impacts the supply chain in terms of cost and quality. Biomass chipped earlier in the supply chain af- fords more flexible and cost-efficient transport options, while chipping later provides more quality control. The survey showed that the majority of chipping takes place in the forest, which suggests the balance of efficiency lies here, while terminal or CHP chipping present logistical challenges in terms of transport utilisation. The location of the chipping process affects the CO2 emissions produced by the wood biofuel supply chains. Chipping at terminals and power plants can be performed by electric chippers, which greatly reduce the CO2 emissions that would be generated if the same bio- mass were chipped in the forest using fossil fuels. This is a trade-off situation between lower costs and lower emissions, as chipping at a terminal/power plant means low emissions but 85 increased transport costs due to power capacity utilisation. Chipping in the forest would re- duce costs by providing better capacity utilisation, but would greatly increase the emissions released in the wood biofuel supply chain. 5.2.5. Overall operation of the supply chain The extensive use of biofuels has been largely attributed to the high carbon tax. The survey revealed that the wood biofuel supply is characterised by uneven demand and depend- ency on buffer stocks and deliveries. Road transport in this regard provides flexibility in com- parison to rail and sea. The respondents in the survey also believe that a sudden change in the policies would present operational difficulties. 5.3. Challenges Explained The issues identified by the literature can be related to the supply chain management interfaces identified by Gold and Seuring (2011) in the bioenergy supply chains. Table 25 provides a summary. The following discussion relates to the logistical challenges identified by the literature review. Table 25: Logistical issues in the SCM steps. “X” denotes the presence of the issue. Logistical issues Op 1 Op 2 Op 3 Op 4 Overall supply system design Difficulties Storage X Location, storage time, estimations, size, un- derutilisation of resources Seasonal varia- bility X Keeping buffer stocks, uneven supply, increased storage costs Chipping pro- cess X X X X Uneven quality, location of the process, handling difficulties Low density of the wood biofu- els X Handling complications, capacity management requirements both in vehicles and at storage ter- minals Term standard- isation X Difficulties in generating statistics Sources of sup- ply X Shortage of wood biofuels Dependency on policies X Change in policies leading to loss of interest in using wood biofuels Seasonal variations in the wood biofuel supply chain present problems with stock planning for energy production. A shortage of biofuels leads to the utilisation of fossil fuels to fill the energy demand, which reduces environment sustainability. Seasonal variations also mean that the demand is high in winter and almost zero in summer. Low demand means un- 86 der-utilization of the equipment with fixed annual costs. Seasonal variations influence the mode of transport used as well, as during high demand seasons rail can be used while, in summer this might not be a viable solution. Seasonal variations affect almost every part of the supply chain. Storage terminals are an integral part of the wood biofuel supply chain, as they help in dealing with seasonal variability. A number of decisions regarding storage such as location, capacity and accessibility of different modes require advance planning. Low demand may lead to under-utilisation of the storage terminals. The chipping process in the supply chain affects the overall costs depending on the lo- cation of the process. Chipped biomass is easier to transport due to improved handling. Chip- ping can be performed at three different locations: forests (roadside), terminals or power plants. The quality of the wood chips produced is also affected by the location, as the chips produced at the terminals and plants are better in quality with more control as compared to the roadside chipping. Roadside chipping, on the other hand, does provide efficient handling in the rest of the supply chain. The location of the chipping process also effects the CO2 emis- sions produced, as forest chipping utilises fossil fuels while a terminal or a plant chipper may be powered by electricity. This presents a trade-off scenario between costs and CO2 emis- sions: forest chipping results in low costs and high emissions while terminal or plant chipping result in higher costs but lower CO2 emissions. The low density of wood biofuels affects the transport cost as well. Logging residues occupy more space and have lower energy content as compared to chipped wood biomass. The use of standardised terms for biofuels is important for reporting purposes, as in any business. A lack of standardised terms can lead to inaccurate reporting of what type of biofuel is needed and in what quantity. This could provide planning problems for the consum- ers (HPs and CHPs). A lack of standardised terms globally would result in a hindrance of trade due to the uncertainty of terms being used for reporting and communication between industries. The successful use of wood biofuels is largely due to European Union policies along with Swedish energy policies. Carbon tax has been the major motivator in this regard for the use of biofuels in Sweden. These environmentally friendly policies together with subsidies helped in discouraging the use of fossil fuels. A sudden change in the policies would present difficulties in keeping biofuels competitive with fossil fuels, as fossil fuels are more efficient in terms of energy density and logistics. The dependence of biofuels on policies makes them 87 vulnerable to abrupt changes. Consistency in Swedish policies, including legislation, taxation, certificate systems, fees and subsidies, has greatly affected the success of biofuels (Björheden et al., 2010). The survey revealed that the logistical challenges could be further narrowed down to several major issues. Table 26 provides a summary of the challenges identified by the survey study. Table 26. Logistical challenges identified from the survey study. The focused and detailed logistical challenges identified by the survey study approve some of the challenges present in the literature, such as seasonal variations, low density of wood biofuels and use of standardised terms. Other challenges such as storage, chipping pro- cess and dependence on policies are described in more detail in the survey study. The storage challenge has been narrowed down to the location and size of the storage facilities. The chip- ping process requires a balance between efficient chipping and transport. Dependence on pol- icies highlights the need for flexibility in the supply chains to respond to changes. Harvesting and collection of biofuels is highly focussed on local sourcing, which limits sourcing options. Regarding the transport of wood biofuels, key challenges are shifting loads from road transport to other transport modes, e.g. intermodal transport along with reduction of environ- mental impact. Operations Current situation Logistical challenges Harvesting and collection Local market focus with truck transport. Local focus limits the possible logistical and sourcing solutions. Storage Essential in the supply chain at both plants and terminals. Balancing size and location against extra costs. Transport Highly dominated by road. The sup- ply chain structure is mainly transport directly from the forests. Finding possibilities for other transport modes to potentially improve the chain, e.g. inter- modal transport. Reducing envi- ronmental impact from road transport. Pre-treatment Most common and preferred loca- tions are the forests and the termi- nals. Balancing between more effi- cient chipping and more effi- cient transport. Overall supply chain Fluctuating demand and buffer stocks. Influenced by political deci- sions. Flexibility in the supply chain to adapt to changes. 88 5.3.1. Sustainability in wood biofuel supply chains Different supply chain options can be developed based on the CO2 emission and cost calculations in the case study of Göteborg Energi plant in the city of Gothenburg. The study demonstrates how economically and environmentally sustainable transport solutions could be developed for a CHP utilising wood biofuels. High utilisation of the resources is key to keep- ing costs down. The case study shows that road distances to terminals should be kept as low as possible, as long haulage distances of 80-100 km will generate higher costs for the system. Rail access both at the plant and near the forest is preferred. The road haulage distance to the receiving terminal is of great importance, as increasing these distances to 100 km will double the costs of the system even if return flows to both trucks and trains are maintained, which is not a realistic option. Biofuels delivered without the use of terminals and road haulage have the greatest potential, for instance products from the forest industry with rail access. Termi- nals have the largest costs, followed by chipping. A large range of terminal costs can be found in the literature and the industry, which makes the selection of the right terminal important. Terminal costs also depend on how costs are shared among different wood products at the terminal, e.g. pulpwood and biofuel. It is important to consider that an all-road solution has many activities in common with intermodal solutions, such as the use of terminals that may also have rail access. The presence of terminal costs in the system makes the additional costs low in an intermodal set- ting where the product has to be loaded onto a train. Therefore, intermodal solutions are com- petitive in comparison to flows already passing through a terminal, especially for round wood, which always passes through a terminal. In an intermodal setting, cost-saving options should be utilised whenever possible, such as low-cost terminal chipping (preferably with an electric chipper) compared to roadside chipping. The price of biofuel also plays a role in the transport cost, e.g. if sourcing from a low-priced region can be done using an intermodal solution while the local all-road sourcing price is expensive and the price difference is significant, the inter- modal solution becomes more attractive. The storage options available at Göteborg Energi contributed to the following conclu- sions about a large HP: x Storage capacity plays an important role in the ordering of shipments. x Missing train deliveries can be managed with a compensating number of trucks, which would lead to less sustainable economic and environmental solutions.. 89 This study addressed many aspects of wood biofuel supply chains, beginning by as- sembling the available literature, collecting market description data, and finishing with a fo- cussed case study. The literature review part of the study provides an introduction notes im- portant aspects of the wood biofuel supply chains. The survey study gathered industry data and lays out the preferences, attitudes and challenges found in the current supply chains. The case study helps in tackling the logistical challenges along with the investigation of potential sustainable intermodal transport options. This study provides a platform for future research endeavours in the field of wood biofuel supply, which can be both specific and general. The sustainability aspect of the study highlights the potential for making wood biofuel supply chains sustainable. This aspect is interesting within the current topic as it is at the heart of utilising wood biofuels and is still open to further exploration. The following section discuss- es some of the future research endeavours that could be attempted based on the current study. 5.4. Future research Some suggestions for future research based on the findings of this study are as fol- lows: 5.4.1. Sustainable international intermodal chains A future research effort could involve an extension of intermodal solutions to an inter- national context. Solutions based upon both rail and sea transportation could be explored. In- volvement of sea transport would provide further investigation in terms of load units, termi- nals, storage handling equipment, etc. 5.4.2. Supply chain development based on fuel type This study focuses on the logistics of wood biofuels. A future study could incorporate different biofuels such as liquid biofuels and note the similarities and differences based on the different biofuels used at the power plants. This would help in giving direction towards spe- cific biofuels that are high in energy content and can be transported efficiently and sustaina- bly. 5.4.3. Development of business models This study outlines the various activities and problems present in the wood biofuel supply chain and presents options for sustainable transport. One future endeavour could be the development of business models that define the practicalities of demand and supply along with contract specifications. These business models could focus on making the business prof- 90 itable without the biofuel-friendly policies, which are currently a cornerstone in the success of wood biofuels. 5.4.4. Social sustainability Further investigation into the evaluation and specification of the social sustainability of the freight transport chains should be developed, as this knowledge can help in better defin- ing socially sustainable supply chains. 5.4.5. GIS based study This study provided a lot of information about the various power plants in Sweden in regards to their operations and locations along with their suppliers and distribution options. The development of a geographical information system for the supply of wood biofuels would help the supply chain managers to properly analyse and manage flows in a more efficient way. A GIS could be used to share the inventory levels of the various storage terminals from which the power plants can source their biofuels. It would also help in planning the shipments required by the power plants, which could mean using fewer vehicles and shifting loads to rail transportation. In addition to these primary benefits, plants would be able to track and monitor their shipments via the GIS, and the system would assist in dealing with the varying demands in a timely manner, thus saving costs. 91 References ABELL, N., SPRINGER, D. W. & KAMATA, A. 2009. Developing and Validating Rapid Assessment Instruments, Oxford Scholarship Online. ALMOTAIRI, B., FLODÉN, J., STEFANSSON, G. & WOXENIUS, J. 2011. Information flows supporting hinterland transportation by rail: Applications in Sweden. Research in Transportation Economics, 15-24. ANAND, S. & SEN, A. 2000. Human Development and Economic Sustainability. World Development, 28, 2029-2049. ANDERSSON, K. 2012. Bionergy the Swedish Experience. Jönköping Sweden: Svebio. ANERUD, E. & JIRJIS, R. 2011. Fuel quality of Norway spruce stumps – influence of harvesting technique and storage method. Scandinavian Journal of Forest Research, 26, 257-266. ASIKAINEN, A. 1998. Chipping terminal logistics. Scandinavian Journal of Forest Research, 13, 386-392. ASMOARP, V. 2013. Terminalstrategier för skogsflis på Södra Skogsenergi (Terminal strategies for wood chips at Södra Skogsenergi). Uppsala: SLU - Swedish University of Agricultural Sciences. BÄCKSTRÖM, S., BOHLIN, M., FRANZEN, U. & JONSSON, P. 2009. Miljökalkyler för intermodala transportkedjor - Detaljerad beräkningsmetodik och relevanta schablonvärden. SIR-C Swedish Intermodal Transport Research Centre. WSP Analys & Strategi. BARBIER, E. B. 1987. The Concept of Sustainable Economic-Development. Environmental Conservation, 14, 101-110. BEHRENDS, S., LINDHOLM, M. & WOXENIUS, J. 2008. The Impact of Urban Freight Transport: A Definition of Sustainability from an Actor's Perspective. Transportation Planning and Technology, 31, 693-713. BEKTAS, T. & CRAINIC, T. G. 2007. A Brief Overview of Intermodal Transportation. Canada: CIRRELT. BERGQVIST, R., FALKEMARK, G. & WOXENIUS, J. 2010. Establishing intermodal terminals. World Review of Intermodal Transportation Research, 3, 285-302. BERGQVIST, R. & FLODÉN, J. 2010. Intermodal road-rail transport in Sweden- the path to sustainability. WCTR. Lisbon, Portugal. BIOENERGIPORTALEN. 2013. Skogsbränslebranschen är ologisk, http://www.bioenergiportalen.se/?p=6630. Bioenergiportalen. BJÖRHEDEN, R., ELIASSON, L., IWARSSON, M., HOFSTEN, H. V., ENSTRÖM, J. & JOHANNESON, T. 2010. Efficient forest fuel supply system. In: THORSEN, Å., BJÖRHEDEN, R. & ELIASSON, L. (eds.). Uppsala: Skogforsk. BLUMBERG, B., COOPER, D. R. & SCHINDLER, P. S. 2005. Business Research Methods, Berkshire, Mcgraw-Hill Education. BOWEN, F., COUSINS, P., LAMMING, R. & FARUK, A. 2001. Horses for courses: explaining the gap between the theory and practice of green supply. Greener Management International Autumn, 41-60. BRADLEY, DIESENREITER, F., WILD, M. & TROMBORG, E. 2009a. World Biofuel Maritime Shipping Study. A study for IEA Bioenergy Task 40. BRADLEY, D., CUYPERS, D. & PELKMANS., L. 2009b. 2nd Generation Biofuels and Trade An exploratory study. IEA Bioenergy Task 40. BRINK, H. I. L. 1993. Validity and reliability in qualitative research. Society of Nurse Researchers' Workshop. South Africa. BUTTON, K. 1993. Transport economics, Aldershot : Edward Elgar. 92 CAMPBELL, D. T. & STANLEY, J. C. 1963. Experimental and quasi-experimental designs for research / Donald T. Campbell, Julian C. Stanley, Boston, Rand McNally, cop. CARGO, G. 2012. Annual report and sustainability report. CARLSSON, D. & RÖNNQVIST, M. 2004. Supply chain management in forestry––case studies at Södra Cell AB. European Journal of Operational Research, 163, 589-616. CARROLL, A. B. 1979. A three-dimensional conceptual model of corporate social performance. Academy of Management Review, 32, 408-23. CARTER, C. R. & EASTON, P. L. 2011. Sustainable supply chain management: evolution and future directions. International Journal of Physical Distribution & Logistics Management, 41, 46-62. CARTER, C. R. & ROGERS, D. S. 2008. A framework of sustainable supply chain management: moving toward new theory. International Journal of Physical Distribution & Logistics Management, 38, 360-387. CHECKLAND, P. 1999. Systems thinking, systems practice - Soft systems methodology - a 30-year retrospective, Chichester, UK, John Wiley. COFORD 2003. Maximising the Potential of Wood use for Energy Generation in Ireland. Horsham: COFORD. CRONBACH, L. J. & MEEHL, P. E. 1955. Construct validity in psychological tests. Psychol Bull, 52, 281-302. DEMPSEY, N., BRAMLEY, G., POWER, S. & BROWN, C. 2011. The Social Dimension of Sustainable Development: Defining Urban Social Sustainability. Sustainable Development, 19, 289-300. DENZIN, N. K. 1970. The research act : a theoretical introduction to sociological methods, London. EC 2007. Presidency conclusions of the Brussels European Council. Brussels: European Council. EKINS, P. 2011. Environmental sustainability: From environmental valuation to the sustainability gap. Progress in Physical Geography, 35, 629-651. ELIASSON, L., GRANLUND, P., JOHANNESON, T., VON HOFSTEN, H. & LUNDSTRÖM, H. 2012. Flisstorlekens effekt på en stor skivhuggs bränsleförbrukning och prestation (Effects of target chip size on performance, fuel consumption and chip quality for a large disc chipper). Uppsala: Skogforsk The Forestry Research Institute of Sweden. ENERGIDATA AS, TRANSPORTØKONOMISK INSTITUTT & CONSULTING, K. 2005. Bioenergy logistics chain cost structure and development potential. Enova SF. ERICSSON, K. 2009. Introduction and development of the Swedish district heating systems. Lund: Lund University. ERICSSON, K., HUTTUNEN, S., NILSSON, L. J. & SVENNINGSSON, P. 2004. Bioenergy policy and market development in Finland and Sweden. Energy policy, 32, 1707-1721. ERIKSSON, T. 2011. Estate and ownership structure [Online]. Online: Swedish Forest Agency. Available: http://www.skogsstyrelsen.se/en/AUTHORITY/Statistics/Subject- Areas/Estate-and-Ownership-Structure/Estate-and-Ownership-Structure/ [Accessed 25-08 2012]. EUROPEAN COMMISSION 2004. Towards a Thematic Strategy on the Urban Environment. Brussels: Commission of the European Communities. FLODÉN, J. 2007. Modelling intermodal freight transport. Doctoral Doctotral Thesis, Göteborg University. FLODÉN, J. 2011. Rail freight costs - Some basic cost estimates for intermodal transport Göteborg: School of Business, Economics and Law at the University of Gothenburg, Swedish Intermodal Reserach Centre, Sir-C, http://hdl.handle.net/2077/25877. 93 FLODÉN, J., BARTHEL, F. & SORKINA, E. 2010. Factors influencing transport buyer's choice of transport service- a european literature review. WCTR. Lisbon Portugal. FORESTRY, S. N. B. O. 2001. Statistical Yearbook of Forestry. Jönköping: Swedish National Board of Forestry. FROSCH, M. & THORÉN, P. 2010. Järnvägstransport av biobränslen. VÄRMEFORSK Service AB. GOLD, S. & SEURING, S. 2011. Supply chain and logistics issues of bio-energy production. Journal of Cleaner Production, 19, 32-42. GOODLAND, R. 1995. The Concept of Environmental Sustainability. Annual Review of Ecology and Systematics, 26, 1-24. GRONALT, M. & RAUCH, P. 2007. Designing a regional forest fuel supply network. Biomass and Bioenergy, 31, 393-402. GUNNARSSON, H., RONNQVIST, M. & LUNDGREN, J. T. 2004. Supply chain modelling of forest fuel. European Journal of Operational Research, 158, 103-123. HAMELINCK, C. N., SUURS, R. A. A. & FAAIJ, A. P. C. 2005. International bioenergy transport costs and energy balance. Biomass & Bioenergy, 29, 114-134. HANKIN, C. A., STOKES, B. & TWADDLE, A. 1995. The transportation of fuelwood from forest to facility. Biomass and Bioenergy, 9, 191-203. HERSLE, D. & BERGLUND, M. 2012. Sustainable Intermodal Supply Systems for Biofuel and Bulk Freight – areas 1.3 and 1.5. WSP. HIRSMARK, J. 2002. Densified Biomass Fuels in Sweden: Country report for the EU/INDEBIF project. Master of Science in Forestry, Swedish University of Agricultural Sciences. HOEFNAGELS, R., RESCH, G., JUNGINGER, M. & FAAIJ, A. 2014. International and domestic uses of solid biofuels under different renewable energy support scenarios in the European Union. Applied Energy, 131, 139–157. HOPWOOD, B., MELLOR, M. & O'BRIEN, G. 2005. Sustainable development: Mapping different approaches. Sustainable Development, 13, 38-52. HUTCHINS, M. J. & SUTHERLAND, J. W. 2008. An exploration of measures of social sustainability and their application to supply chain decisions. Journal of Cleaner Production, 16, 1688-1698. JANIC, M. 2006. Sustainable transport in the European Union: A review of the past research and future ideas. Transport Reviews, 26, 81-104. JENSEN, A. 1990. Combined transport - systems, economics, and strategies, Stockholm, Swedish Transport Research Board (TFB). JOHANSSON, J., LISS, J. E., GULLBERG, T. & BJÖRHEDEN, R. 2006. Transport and handling of forest energy bundles—advantages and problems. Biomass and Bioenergy, 30, 334–341. JOHANSSON, J. & MORTAZAVI, R. 2011. Transport av skogsenergisortiment - Företags- och samhällsekonomiska kostnader. SWX-Energi. JOURQUIN, B., BEUTHE, M. & DEMILIE, C. L. 1999. FREIGHT BUNDLING NETWORK MODELS: METHODOLOGY AND APPLICA TION. Transportation Planning and Technology, 23, 157-177. JUNGINGERA, M., BOLKESJØB, T., BRADLEYC, D., DOLZAND, P., FAAIJA, A., HEINIMO, J., HEKTORF, B., LEISTADG, Ø., LINGH, E., PERRYI, M., PIACENTED, E., ROSILLO-CALLEI, F., RYCKMANSJ, Y., SCHOUWENBERGK, P.-P., SOLBERGL, B., TRØMBORGL, E., WALTERD, A. D. S. & WITA, M. D. 2008. Developments in international bioenergy trade. BIOMASS AND BIOENERGY, 32, 717-729. 94 KIRK, J. & MILLER, M. L. 1986. Reliability and Validity in Qualitative Research, Newbury Park, SAGE Publications, Inc. KONINGS, J. W. 1996. Integrated centres for the transshipment, storage, collection and distribution of goods. Transport Policy, 3, 3-11. KROSNICK, J. A. 1999. Survey research. Annu Rev Psychol, 50, 537-67. LABUSCHAGNE, C., BRENT, A. C. & VAN ERCK, R. P. G. 2005. Assessing the sustainability performances of industries. Journal of Cleaner Production, 13, 373-385. LAMERSA, P., JUNGINGER, M., HAMELINCK, C. & FAAIJ, A. 2012. Developments in international solid biofuel trade—An analysis of volumes, policies, and market factors. Renewable and Sustainable Energy Reviews, 16, 3176–3199. LAMMGÅRD, C. 2007. Environmental Perspectives on Marketing of Freight Transports. Doctorate, Univeristy of Gothenburg. LARSSON, F. & NYLINDER, M. 2014. WeCalc - Wood Energy Calculations [Online]. Uppsala: SLU - Swedish University of Agricultural Sciences. Available: http://woodenergy.sites.djangoeurope.com/conversion/ [Accessed March 2014]. LECOMPTE, M. D. & GOETZ, J. P. 1982. Problems of Reliability and Validity in Ethnographic Research. Review of Educational Research, 52, 31-60. LINDHOLM, E. L. & BERG, S. 2005. Energy requirement and environmental impact in timber transport. Scandinavian Journal of Forest Research, 20, 184-191. LITTIG, B. & GRIESSLE 2005. Social sustainability: a catchword between political pragmatism and social theory. Int. J. Sustainable Development, 8, 65-79. LOMBARDINI, C., GRANLUND, P. & ELIASSON, L. 2013. Bruks 806 STC – Prestation och bränsleförbrukning (Performance and fuel consumption of the Bruks 806 STC chipper). Uppsala: Skogforsk The Forestry Research Institute of Sweden. LOWE, D. 2006. Intermodal Freight Transport. Taylor and Francis. LUMBRERAS, J., BORGE, R., GUIJARRO, A., LOPEZ, J. M. & RODRÍGUEZ, M. E. 2013. A methodology to compute emission projections from road transport (EmiTRANS). Technological Forecasting & Social Change, 81, 165-176. MAHMUDI, H. & FLYNN, P. C. 2006. Rail vs truck transport of biomass. Appl Biochem Biotechnol, 129-132, 88-103. MCKINNON, A., BROWNE, M. & WHITEING, A. 2012. Green Logistics, United kingdom, Kogan Page Limited. MEYER, P. 2010. Understanding Measurement: Reliability, Oxford Scholarship Online. MIAO, Z., SHASTRI, Y., GRIFT, T. E., HANSEN, A. C. & TING, K. C. 2012. Lignocellulosic biomass feedstock transportation alternatives, logistics, equipment configurations, and modeling. Biofuels, Bioproducts and Biorefining, 6, 351-362. MIKKILÄ, M., HEINIMÖ, J., PANAPANAAN, V., LINNANEN, L. & FAAIJ., A. 2009. Evaluation of Sustainability schemes for international bioenergy flows. International journal of energy sector management, 3, 359-382. MIN, H. & KIM, I. 2012. Green supply chain research: past, present, and future. Logistics Research, 4, 39-47. MOLDAN, B., JANOUSKOVA, S. & HAK, T. 2012. How to understand and measure environmental sustainability: Indicators and targets. Ecological Indicators, 17, 4-13. NELLDAL, B. L., BARK, P., ANDERSSON, E., TROCHE, G., ÖSTLUND, S., JÖNSSON, P. A., STICHEL, S., WENNHAGE, P., CARLSSON, U., SUNDQVIST, H. K., JAMES, G., BERGSTEDT, R. & WAJSMAN, J. 2005. Efficient train systems for freight transport A systems study. In: NELLDAL, B.-L. (ed.) KTH Railway Group Report Stockholm. NOZICK, L. K. & MORLOK, E. K. 1997. A model for medium-term operations planning in an intermodal rail-truck service. Transpn Rev, 31, 91-107. 95 NUNNALLY, J. 1959. Tests and Measurements, New York, McGraw-Hill. OECD 2001. OECD Environmental Strategy for the First Decade of the 21st Century. Organization for economic cooperation and development. OLSSON, O. 2006. The Swedish biofuel market -studies of Swedish foreign biofuel trade and of the consequences of hurricane Gudrun. Uppsala Universitet. PARK, C. 2007. A Dictionary of Environment and Conversation. Oxford University Press. PRIEMUS, H., NIJKAMP, P. & SHEFER, D. 1999. Intermodality and sustainable freight transport. Transportation Planning and Technology, 23, 79-81. RAO, P. & HOLT, D. 2005. Do green supply chains lead to competitiveness and economic performance? International Journal of Operations & Production Management, 25, 898-916. RAUCH, P. & GRONALT, M. 2003. The terminal location problem in the forest fuels supply network. International Journal of Forest Engineering, 21, 32-40. RAUCH, P. & GRONALT, M. 2010. The effects of rising energy costs and transportation mode mix on forest fuel procurement costs. Biomass and Bioenergy, 35, 690-699. REIS, V., MEIER, J. F., PACE, G. & PALACIN, R. 2013. Rail and multi-modal transport. Research in Transportation Economics, 41, 17-30. RENTIZELAS, A. A., TOLIS, A. J. & TATSIOPOULOS, I. P. 2009. Logistics issues of biomass: The storage problem and the multi-biomass supply chain. Renewable & Sustainable Energy Reviews, 13, 887-894. ROOS, A., BOHLIN, F., HEKTOR, B., HILLRING, B. & PARIKKA, M. 2000. A geographical analysis of the Swedish woodfuel market. Scandinavian Journal of Forest Research, 15, 112-118. ROUTA, J., ASIKAINEN, A., BJÖRHEDEN, R., LAITILA, J. & RÖSER, D. 2012. Forest energy procurement: state of the art in Finland and Sweden. WIRES Energy Environment. SARKIS, J. 1995. Supply chain management and environmentally conscious design and manufacturing. International Journal of Environmentally Conscious Design and Manufacturing, 4, 43-52. SAVITZ, A. W. & WEBER, K. 2006. The tripple bottom line, San Francisco, CA, Jossey- Bass. SAXIN, B., LAMMGÅRD, C. & FLODÉN, J. Meeting the demand of goods transport - identification of needs among Swedish companies. NOFOMA 2004, 7-8 june 2004 Linköping, Sweden. SELKIMAKI, M., MOLA-YUDEGO, B., ROSER, D., PRINZ, R. & SIKANEN, L. 2010. Present and future trends in pellet markets, raw materials, and supply logistics in Sweden and Finland. Renewable & Sustainable Energy Reviews, 14, 3068-3075. SEURING, S. & MÜLLER, M. 2008. From a literature review to a conceptual framework for sustainable supply chain management. Journal of Cleaner Production, 16, 1699-1710. SKOGFORSK 2011. FLIS 4.0 Flexibel lathund för interaktiv systemanalys. Uppsala: Skogforsk The Forestry Research Institute of Sweden. SOMMAR, R. 2010. Utvärdering av intermodala transportkedjor - kostnadsmodeller. TFK, KTH, SIR-C Swedish Intermodal Transport Research Centre,. STOCKARYDSTERMINALEN. 2009. Stockarydsterminalen [Online]. Available: http://www.stockarydsterminalen.se [Accessed 23-07 2011]. SUTTON, P. 2004. A Perspective on environmental sustainability? A paper for the Victorian Commissioner for Environmental Sustainability. Available: http://www.green- innovations.asn.au/A-Perspective-on-Environmental-Sustainability.pdf [Accessed 13- 04-2013]. 96 SVANBERG, M. & HALLDÓRSSON, Á. 2013. Supply chain configuration for biomass-to- energy: the case of torrefaction. International Journal of Energy Sector Management, 7, 65-83. SVENSK FJÄRRVÄRME. 2010. The Swedish District Heating Association Homepage [Online]. Available: http://www.svenskfjarrvarme.se/Statistik-- Pris/Fjarrvarme/Energitillforsel/ [Accessed 2012 January 6. SWEDISH DISTRICT HEATING ASSOCIATION. 2011. The Swedish District Heating Association [Online]. Available: http://www.svenskfjarrvarme.se/Statistik-- Pris/Fjarrvarme/Energitillforsel/ [Accessed 01-06 2012]. TAHVANAINEN, T. & ANTTILA, P. 2011. Supply chain cost analysis of long-distance transportation of energy wood in Finland. Biomass and Bioenergy, 35, 3360-3375. THRAEN, D., TROSSERO, M. A., DRIGO, R., BOURKE, J. & NOGUEIRA, L. H. 2004. Unified bioenergy terminology. In: TINA ETHERINGTON, ROBERT FLOOD, MARCO PERRI & GUARDIA, M. (eds.). Food and agriculture organization of the United Nations. TIMILSINA, G. R. & SHRESTHA, A. 2011. How much hope should we have for biofuels? Energy, 36, 2055-2069. TRAD, S. 2010. The Swedish District Heating Association [Online]. Available: http://www.svenskfjarrvarme.se/Statistik--Pris/Fjarrvarme/Energitillforsel/ [Accessed 01-06 2012]. TRAFIKVERKET 2010a. Pocket Facts 2010 – The Swedish Transport Administration, railways, roads, traffic and transports. In: LUNDSTRÖM, B. (ed.). Borlänge: The Swedish Transport Administration. TRAFIKVERKET 2010b. The Swedish transport administration annual report 2010. Borlänge: Swedish transport administration. UNITED NATIONS 1980. Convention on International Multimodal Transport. New York, United States of America: United Nations. UNITED NATIONS 2001. Terminology on combined transport. Geneva, Switzerland. WIEDMANN, T. & MINX, J. 2008. A Definition of Carbon Footprint. Ecological Economics Research Trends. Newyork USA.: Nova Science Publishers, Hauppauge NY, USA. WIEGMANS, B. W., MASUREL, E. & NIJKAMP, P. 1999. Intermodal freight terminals: an analysis of the terminal market. Transportation Planning and Technology, 23, 105- 128. WOLFSMAYR, U. J. & RAUCH, P. 2013. The primary forest fuel supply chain: A literature review. Biomass and bioenergy, 60, 203-221. WORLD COMMISSION ON ENVIRONMENT AND DEVELOPMENT 1987. Our common future, Oxford ; New York, Oxford University Press. YIN, R. K. 2009. Case study research: Design and methods, London, England, Sage. ZHOU, Z. Y., CHENG, S. W. & HUA, B. 2000. Supply chain optimization of continuous process industries with sustainability considerations. Computers & Chemical Engineering, 24, 1151-1158. 97 Appendices Survey 98 99 100 101 102 103 104 105 106 107 108 109 110 Cost Data Input data Density kg/loose m3: Bark 350, Log chips 271, Logging residue chips 295, Saw residue chips 300, Sawdust 30, Stump chips 288, Whole three chips 300, Wood residue chips 225, Logg 367 Energy MWH/m3: Bark 0.65, Log chips 0.79, Logging residue chips 0.78, Saw residue chips 0.65, Sawdust 0.58 Stump chips 0.77, Whole three chips 0.8, Wood residue chips 0.8, Logg 1.07 Truck cost considering empty return, kr/km: Wood chip container truck 29.57 kr, Wood chip truck 30.77 kr, Forest residue truck 34.45 kr, Timber truck 29.57 kr Truck emissions considering empty return, CO2 kg/km: Wood chip container truck 2.95, Wood chip truck 3.02, Forest residue truck 3.13, Timber truck 2.24 Rail engine annual fixed costs: Rd 2 149 868 kr, Modern engine 3 426 462 kr Rail engine variable costs, kr/km: Rd 16.44 kr, Modern engine 19.08 kr Rail staff cost: 692,79 kr/train hour Wagon annual fixed costs: Lgns 41 221 kr, Sgns 63 119 kr Wagon variable costs, kr/km: Lgns 0.23 kr, Sgns 0.31 kr Load unit annual costs: Rotary container 13 100 kr, 40m3 container 11 521 kr Terminal costs, kr/MWh: Road to road 14 kr, Road to rail 19 kr, Rail to road 21 kr Fork lift truck annual fixed costs: Heavy truck with rotator 764 141 kr, Light truck 192 920 kr Fork lift truck staff cost: 310 kr/hour Chipping cost, logging residues, kr/MWh: Mobile chipper at terminal 23 kr, stationary electric chipper 7 kr, roadside chipping 40 kr Chipping cost, Logg, kr/MWh: Mobile chipper at terminal 20 kr, stationary electric chipper 5 kr, roadside chipping 35 kr 111 Paper 1 Awais, F., 2013, Wood biofuels logistical challenges in Sweden. Presented at the NOFOMA conference 2013, Gothenburg, Sweden. Wood biofuels logistical challenges in Sweden. Fawad Awais*. *Department of Business Administration at University of Gothenburg, Vasagatan 1, Hus D, plan 6, Box 610, SE 405 30 Gothenburg, Sweden. E-mail: fawad.awais@handels.gu.se , Tel: +46 31 - 786 2617 Fax: 031 - 786 1466 Website: http://www.fek.handels.gu.se/kontakt/medarbetare/fawad-awais/ ABSTRACT Purpose The basic purpose of the paper is to review the current literature on the subjects of wood bio- fuels and identify key logistical challenges for different types of biofuel. Approach The paper involved an investigation about the definitions used for the wood biofuels and their raw materials in the research of the logistics of wood biofuels. The generally used distribution networks are identified and analysed. The paper is then concluded with the various logistical challenges that are present in the wood biofuel industry. Information and data are obtained through a literature review study. Findings This paper gives an introduction to the definitions for wood biofuels and its raw materials. It also outlines the various logistical problems that arise in the distribution of wood biofuels and its raw materials for use in heating and power plants in Sweden. Key identified challenges are seasonal variations, storage, chipping process, the low density of the wood biofuels, term standardisation, sources of supply and dependency on policies. Research limitations Transport of wood chips and wood pellets have been considered along with their raw materi- als and sources. The focussed supply chains include chains from domestic forests and import- ed goods. A call for further investigation to develop sustainable wood biofuel supply chains is made in the conclusions. Originality/value This paper contributes by analysing the supply chain for wood biofuel from a holistic perspec- tive and giving a solid foundation for further research into how these challenges can be solved. Keywords: Wood, biofuel, transport, logistics, challenges. 112 INTRODUCTION In the year 2010, biofuels, peat and waste amounted to 135 terawatt hour (TWh) which amounts to almost 22 % of the total energy supplied to Sweden, whereas nuclear energy was 166 TWh which is 27% of the total energy supplied (The Swedish Energy Agency, 2012). The Swedish Commission on Oil Independence has estimated the supply of biofuels to Swe- den to increase to 154 TWh/year by the year 2020 (Eriksson, 2008). The concept of biofuels involves a diverse range of products and substances, which are utilised in today’s energy de- prived world. Biofuel in general terms can be defined as any fuel that is derived from bio- mass. Wood from trees form one of the largely used solid biofuels and can be transformed into products such as wood pellets, torrefied wood and bio carbon (Bradley et al., 2009b). This study utilised the already existing literature which is present digitally from journal data- bases along with research reports from other institutions in Sweden. The literature reviewed was focussed on the transportation of wood biofuels on different modes. An effort to provide comprehensive results with a focus of domestic wood biofuel transportation has been made. The topic of biofuels includes different types of biofuels along with issues not only in the field of logistics but also other technical fields. Therefore, defining keywords and selecting specific databases was not an easy task, along with the special focus on the domestic wood biofuel transportation. However keywords that were used to search the material studied were “wood fuels transportation”, “wood biofuels transportation”, “biofuel supply chains” and “is- sues in biofuel supply chains”. The databases that were extensively used were “Science Di- rect” and “Emerald” along with “Google Scholar”. Since the databases showed different re- sults along with studies related to other geographical locations, a lot of valuable insights were provided by the other research reports present online from various institutions and organisa- tions such as the Swedish University of Agricultural Sciences and The Swedish Energy Agen- cy and Statistics Sweden. With fossil fuels widely accepted as non-renewable sources of energy, attention has been di- verted to alternative sources, which are more environmentally sustainable. Insecurity of the environment and the supply of fossil fuels are the main drivers behind the current interest in forest fuels. Europe, for instance has diverted its focus to forest fuel in order to find fuels that are not only renewable but are also CO2 neutral.(Rauch and Gronalt, 2010). Climate persever- ance is at the heart of the concept of forest fuels. With the growing demands of forest fuels, development of a cost efficient distribution network is considered a logical step. This study helps in compiling the logistical issues with a focus on the domestic transportation of wood biofuels which has been rarely reported in the existing literature, specifically. Academically this study provides a compilation of the various logistical challenges present in the supply chain of wood biofuels and calls logisticians for the investigation of sustainable transportation solutions. Empirically this paper contributes by identifying the major challenges, which can help in creating better supply chain designs. This paper aims at outlining the various logistical problems that arise in the distribution of wood biofuels and its raw materials to heating plants (HP) and combined heat and power plants (CHP) in Sweden. The focus is on wood biofuels used in CHPs, which are mainly wood chips and wood pellets. This paper starts with the introduction of the wood biofuels under consideration. It moves further with the logistical activities involved in the transportation of wood biofuels followed by the design of the supply chain. The next section discusses the logistical challenges present in the wood biofuel supply chains, which is followed by an analysis of these challenges. The paper is finally concluded with the main findings of the study followed by acknowledgements and a list of references. 113 Wood biofuels Wood chips are wood that has been chipped into small pieces. It is comprised of a mixture of hard and soft wood that are reduced to a size of approximately 5-8 centimetres with heteroge- neous shape (Bradley et al., 2009a). Wood pellets are densified biofuels, which are made from pulverised biomass that are com- press into a 5-30mm cylinder. According to the Swedish standards, wood pellets are cylindri- cal in shape with a diameter of a maximum of 25mm. The raw materials used for these pellets are saw dust, shavings, chippings and by-products from forests or agriculture operations. (Hirsmark, 2002). LOGISTICS IN WOOD BIOFUEL TRANSPORTATION In the 1970s the interest of biofuels to be used as an alternative to fossil fuel emerged due to the oil crisis (Timilsina and Shrestha, 2011). District heating plants are one of the major consumers of biofuels (natural gas, wood biofuels, energy crops, etc.), and the extent of consumption has increased over the years, the Swedish district heating sector being particularly rich in wood and having shown a visible increase in the use of wood biofuels over the course of time, see figure 3.1. Distict heating sector is comprised of heating plants (HP) and combined heating plants (CHP). Both categories produce heat energy except CHPs, which produce electricity as an additional product. The increase in the use of biofuels is mainly due to the increase in carbon tax along with industry implementing measures to reduce the use of fossil fuels. This has led to major reductions in carbon emissions and increases in the use of biofuels. However, this increasing use of biofuel creates new logistical challenges, as the huge volumes need to be transported to energy plants. At present, Sweden hold adequate production of the wood biofuels through the use of residues from wood processing industry and forests for the HPs and CHPs. Use of long distance transports with large volumes is considered feasible along with being environmental- ly attractive. However, the transportation chains of the wood biofuels faces many challenges on the basis of costs, energy use, material loss and applied logistics.(Eriksson, 2008). Gold and Seuring (2011) has considered a wide variety of bio fuels but the statistics of the papers reviewed by them show a major representation of the wood biofuels. They argue that 0 5 10 15 20 25 30 35 1970 1980 1990 2000 2010 2020 Figure 3.2 Use of wood biofuels in district heating plants, 1980–2010, in TWh (Source: Swedish Energy Agen- 114 there are two major objectives of a biomass supply, which are that costs of wood biomass should be kept competitive and that wood biomass should be supplied continuously. In this regard problems may occur due to the growing cycles of most biomass, minimal co- ordination, reliability and willingness of the supply chain actors. The main operations that are present in the supply chains of the biofuels have been outlined in the table 3.1. Table 3.3 Various steps of SCM/logistics and bio-energy. Adapted from:(Gold and Seuring, 2011). Category Description Collecting/ harvesting This part of the supply chain takes place in the forest where the wood is collected and is more focussed on forest management rather than logistics. The chipping process taking place in the forest affects the handling and costs of the wood biofuel transportation. Storage This is a key step in the supply chain as it helps when encountering seasonal variations for the demand and supply of wood biofuels, along with providing another point for the chipping process. Transport This is involved in between various points of the wood biofuel supply chain such as the forests, storage terminals, heating plants, etc. The form of wood biofuels along with the mode used for transportation are the key areas in this step. Pre-treatment activities This involves chipping, drying and conversion of wood biomass to densified biofuels such as wood pellets. These activities have economic, environmental and social impacts on the other operations of the biofuel supply chain. Overall supply system design This refers to the challenging task of effectively and efficiently designing and operating bio-energy production. Most of the literature focuses on the overall layouts of the biomass supply chains rather than the isolated components. The first focused area in the biomass supply is the supply chain ar- chitecture, which involves optimisation solutions for the location of storage and chipping pro- cess. Biofuel chains involve complexity along with different actors and market segments. The energy plants and the biomass suppliers have been identified as the two most important actors in the biomass supply chain. These two actors are interdependent in a sense as the energy plants need biofuels and the producers need the energy plants, which utilise biofuels. The en- ergy plants that utilise biofuels are small as compared to the fossil fuel based plants, but the logistics are really complex as the biofuel based energy plants require more deliveries due to low energy content, since the same quantities of fossil fuels yield more energy in comparison to wood biofuels (Gold and Seuring, 2011). Consider 1 TWh which is equivalent to 86,000 tonnes of equivalent of oil. One tonne of wood pellets hold 4.6 megawatt hour (MWh) of en- ergy which means 1 TWh corresponds to 217,000 tonnes of wood pellets (Svebio and Andersson, 2012). Due to the high energy content of fossil fuels, it requires a lower number of delivery trips to the energy plants to produce the same amount of energy as the wood biofuels. The biofuel plants require low transportation costs along with the availability of more biomass 115 for the plants to operate at an optimum level. One of the main methods to achieve optimum levels in biomass supply chain management is the supply chain architecture, which is charac- terised by the concept of economy of scale. Economy of scale is an incentive for large scale operations for the biofuel based plants along with the development for utilising other biofuels. The other main method includes tools for enhancing supply chain functioning: Logistical ac- tivities involved in the biomass supply chain are thought to be critical thus logistical efficien- cy is usually tried to achieve through the use of software and analysis tools (Gold and Seuring, 2011). Gunnarsson et al. (2004) discuss the wood biofuel supply chain presents a very original sup- ply chain problem. Companies that supply (forest owners, wood processing industries, trans- porting companies) and consume wood biofuels (HPs and CHPs) are faced with different sce- narios regarding their supply chain. Some of the scenarios involved in the wood biofuel chains are which fuel type to be used, the time required for the movement of the wood biofu- els, whether or not to employ the process of chipping, location of the chipping process, stor- age at terminals, design of transportation network, establishing contracts with the forest and saw mill owners and restrictions regarding capacities of chipping, forwarding and storage at terminals. Gunnarsson et al. (2004) have also identified key strategic planning situations for the supplying companies involved in the wood biofuel chains, which are submitting competi- tive prices, finding a solution to the varying demands of the wood biofuels—e.g. in colder than usual months—adopting a new storage terminal or altering the capacity of the already existing storage terminal, altering the capacities of the chipping sites and the modes of trans- portation, changing costs due to the change in the chipping technology and negotiating with the transporting companies. Gunnarsson et al. (2004) focussed on the planning levels for the wood biofuel supply chain where key choices define the design of the supply chains. They also highlight the activities in a wood biofuel chain that are interesting to study in the field of logistics and how modelling of such a chain can be performed. A normal biomass chain can be said to have harvesting and collection, storage, transport, and pre-treatment techniques as the main operations involved. Gold and Seuring (2011) reveal that the overall design of the biomass chains is the most focussed area of the literature reviewed followed by the harvesting and the collection process. The topics with less focus are the stor- age and the pre-treatment techniques of the biomass. Harvesting and collecting biomass The first operation is to harvest and collect the biofuel in the forest. This occurs with special- ised forest machinery and is outside the scope of this paper. However it is important to note that this operation may or may not involve the chipping process, which has direct cost impli- cations on the whole supply chain of the wood biofuels (Gold and Seuring, 2011). Figure 3.1.1 Chipping process. 116 The transportation of the logging residues and the chipping process are the major factors of cost in the production of wood biofuels. Decisions about the place of chipping and technology employed can affect the costs of the overall chain. Use of the new efficient technology and methods can mean lower costs e.g. in the case of loose residues, utilisation of trucks with mounted wood chipping equipment can reduce the costs by 20-40% in small logging sites. Over the large sites, chipper trucks can be competitive for a travelling distance of 100-120 km. (Björheden et al., 2010). Routa et al. (2012) describe that the wood biofuel supply chains are often built in accordance to the chip- ping of wood biomass. The place of the chipper or the crusher in the supply chain defines the physical form of the wood biomass in the transportation process. The chipping of the wood biomass is commonly performed at three different locations, which are the forests, storage terminals or the plants. Storage The major reason that storage is involved in the biomass chains is to maintain a smooth sup- ply to the energy plants as per their demand. The short period of the harvesting time and the scattered locations of the biomass make storage a necessity for the biomass chain. Generally the shorter the period of harvest, the higher the number of storage terminals are needed as a buffer for the smooth supply to the energy plants. Therefore storage terminals are established throughout the biomass chain. Covered storage terminals can serve as a place for drying the biomass along with the storage capability (Gold and Seuring, 2011). An example of storage would be the stocks of the extracted wood kept along the roadside. Storage is also part of the drying process when wood biofuels are left to dry in order to reduce the moisture level from 50% to 30% without bearing any costs. Storing wood materials in the open air causes a 3% dry mass loss due to decomposition. Logs and wood chips are often stored outside. Processed wood biofuels such as dried chips or pellets, which are more valua- ble, are kept inside bunkers for protection. Keeping processed wood biofuels inside is of great advantage as storage areas such as bunkers and silos have no negative effect on the wood bio- fuel (Hamelinck et al., 2005). Transport in the bio-energy chain The transport issues of the biomass chains have been divided into 3 categories. The first one is legal and infrastructural framework. The second category includes main variables impacting transport operations, such as transport time, which is influenced by the distance and speed of Figure 3.1.2 A chip truck equipped with a chipper and a hydraulic crane. 117 the transportation along with the lower energy density of the biomass being transported. The third category includes social and environmental impacts of the transport; transportation of the biomass have environmental aspects as well, such as the CO2 emissions along with the societal impacts such as traffic congestion(Gold and Seuring, 2011). Road transportation is a key part of the transport in wood biofuel supply chains. Road trans- portation is involved in delivering wood biomass to the energy plants directly if the transport distances are small. For longer distances trucks are involved in the initial and final haul of the wood biomass except for facilities that are equipped with rail sidings or ship harbours. Johansson et al. (2006) describe the typical road vehicles used in Sweden with a maximum length and width of 24 and 2.60 meters respectively with an allowable carrying weight of 60 tonnes. In their study they describe the comparison of transport of bundled and chipped wood biomass. Bundling has been said to improve wood biofuel logistics in which uniform handling units are created, which allow utilisation of full capacity of the vehicles. The bundling system is currently overshadowed by the dominant chipping system however Johansson et al. (2006) described some scenarios when it is more cost efficient, but the system is considered to be in the experimental phase. Large consumers (CHPs and HPs), who have problems resourcing their demand locally con- sider rail and sea transports important for the supply networks. Sweden today has wood chips being transported in containers through rail transportation as well. Economical, long distance transportation can balance out the prices for forest fuels, while larger regions, which may have multiple extraction sites, can benefit from the presence of train transportation. There is a need to increase the efficiency of the rail transportation in order to increase its usage for transporting wood biofuels. Estimation of the train transports is considered difficult as the costs vary considerably in a wide variety of the conditions. The key factors that influence the costs for the rail transportation of wood biofuels are deliveries made by the train units per unit time, utilisation of load capacity, transport distance, storage terminal handling (loading and unloading) and conditions of train shunting at storage terminal points (Björheden et al., 2010). High fixed costs are associated with train transport which can vary between 50-60 per cent. Therefore large volumes are required to cover the fixed costs along with full load capacity. Figure 3 shows that transport distance has direct effects on the variable costs for train transport, but this also influences the number of deliveries made per week. Handling opera- Figure 3.3.1 Examples of costs for railway transport of chips over different distanc- es and numbers of deliveries per week SEK/MWh (Björheden et al., 2010). 118 tions are also important as the time and equipment utilised at a storage terminal incurs costs. In train transport the goods in most cases cannot be unloaded directly at the plants due to lack of rail access. Figure 3.3.2 Innofreight container with forklift (source: www.innofreight.com) The common systems for the transport and loading and unloading processes are Innofreight and the container system for switch body trucks. In the Innofreight system, large containers (46 or 53 m3) are utilised and emptied by rotating them with a special forklift truck with a turntable. This system is usually used at plants that receive several deliveries throughout the week by rail transportation. Another is the container system for switch body trucks. This sys- tem involves containers that can be carried on both the trucks and trains. There is space pro- vided under the containers for the forklifts to grab the containers. These are unloaded from trains through the use of forklifts, and the truck empties them by tipping them back while pulling through a hook. These containers can also be pulled onto a trailer if they are required to be transported before they are to be emptied (Björheden et al., 2010). The two systems are shown in Table 3.3.1. Table 3.3.1 Main loading and unloading techniques of wood bio- fuels (Björheden et al., 2010). Systems Advantages Disadvantages The Innofreight sys- tem Specially designed containers mean greater volumes on each wagon. Efficient unloading with turn- table forks The unloading forklift truck is special- ised and has few other uses. The forklifts are designed for Innofreight containers and cannot handle other containers. Container system for switch body trucks Long distance transportation of the containers by trucks can be done as containers are designed for both trucks and trains. Includes standard components. Longer loading and unloading time is required as compared to the Innofreight system. Load capacity cannot be utilised to the maximum of the train as the weight and width limit of the containers on trucks and trains are different, with trains having wider limits. 119 Pre-treatment techniques The two most common pre-treatment techniques employed to the biomass are drying and Pelletisation process. The drying process increases the combustion efficiency of the biomass along with its improved resistance to decomposition and fire hazards but these advantages come with added costs. The drying process also improves the handling of the material as the material becomes lighter due to the absences of moisture. Drying in the open air reduces the cost to zero, but this is highly dependent on the climatic conditions of a region. Pelletisation is a pre-combustion treatment, which reduces the moisture content and increases the bulk densi- ty of the biomass, thus wood pellets are considered to hold higher levels of energy content and are able to directly substitute coal. Other advantages include improved handling and transpor- tation efficiency. Pelletisation can also withstand prolonged storage time periods without sig- nificant dry matter losses. Pelletisation comes with higher costs as compared to the other me- chanical solutions as it requires producing a densified product with more energy content as compared to ordinary wood chips (Gold and Seuring, 2011). SUPPLY SYSTEM DESIGN General representations of the transportation activities are presented in figure 4.1. 120 The arrows in the diagram represent the flow of the wood biofuel raw materials and products. Each flow represents a different combination of the flow of wood chips and pellets. The trian- gles in the diagram represent the onsite chipping process e.g. the triangles adjacent to the for- ests represent the road side chipping done in the forest. The logistical activities that take place between the origins and destination of the wood biofuels are elaborated by the two main activ- ities, which are the transport and the loading/unloading intermediary activities. It should be noted that transports are carried directly from the location of origin to the power plants along with passing through storage terminals and the pellet production plants. The actors in the chains are significant to consider as they influence the supply of the raw materials for the wood biofuel industry. LOGISTICAL CHALLENGES Logistically wood is heavy to transport and provides less energy as compared to fossil fuels. This implies an aspect of being economical for the transport of wood biofuels (Rauch and Gronalt, 2010). A significant amount of the costs of generating energy by biofuels is due to the logistical activities involved in the process. Storage is said to be one of the major con- cerns, which is highly characterised by the seasonal variations of the biofuels (Rentizelas et al., 2008). Another issue in the transportation of wood biofuels is the low density of the goods being transported. Finally, the different forms of biomass put requirements of different Figure 4.1 National supply chains under focus. Triangles represent the onsite chipping. Source: Self-drawn. 1* Operation 1: harvesting and collecting biomass. 2* Operation 2: Storage. 3* Operation 3: transport in the bio-energy chain. 4* Operation 4: pre-treatment techniques. 121 equipment for handling purposes e.g. chips and logs require different handling equipment (Rentizelas et al., 2008). Similar and many other such problems require close attention to the distribution process of the wood biofuels in order to compete economically with fossil fuels. Seasonal variations Seasonal variations have an effect on the demand of the heat energy as well. Naturally, more energy is needed in the winters as compared to summers. This creates a need for an accurate forecast of the wood biofuels for the different seasons. Severity of the weather can also be an issue in this regard, as a colder winter would require more energy distribution which in turn requires more fuel. In the wood biofuel chains the stocks vary to a great degree not because of the market but due to the seasonal variations of the forest management. In Sweden, traditionally the harvesting period of the wood biofuels is during autumn, winter and early spring, which is evident in the overall wood flow in the chain. As a result there is a large supply of wood during the spring and very few during the summer. The critical period, in this regard, is just after summer, when the various actors in the supply chain plan their demand for wood biofuels. The general prob- lem is to have enough stocked for the month of August and early September, after which the harvesting processes again begin after the vacation period (Carlsson and Rönnqvist, 2004). Seasonal variations of wood biofuels also results in the underutilisation of the expensive ma- chinery and equipment, which results in the increase of the annual operational costs. Addi- tionally, annual harvesting of the wood biofuels creates more labour costs in comparison to the perpetual harvest. Furthermore the shorter the harvest periods, the greater the period of stock retention, resulting in higher costs of storage and dry matter losses (Gold and Seuring (2011). This means that adequate buffer stocks are needed to be kept in the inventory for a smooth flow of raw materials. Storage terminals become very important in this regard as some power plants lack the large storage facilities. Utilisation of different modes can differ as well in the different seasons. Large volumes in the winters mean utilisation of rail transportation which cannot be utilised in summers when the demand is low. Storage Due to the uneven demand and supply of the wood resources from forests, storage has be- come an integral part of the supply chain. At different levels of the chain, storage is needed to fill the gaps between the supply and demand. The necessity of the storage facility for the wood biofuel chains involves a number of logistical decisions regarding the location, size and equipment present at a storage facility. Storage facilities deserve special attention as they di- rectly affect the overall cost of the supply chain. Chipping process Reducing the extracted wood to a certain size is also a concern that is part of the supply chain. It presents a decision-making scenario regarding the location where the sizing of the wood materials should be performed. Normally, three types of choices are available regarding the sizing issue, which are chipping at the roadside, at the storage terminal or pellet production plant and at the energy plants (HPs & CHPs). The chipping applied to the wood materials varies as different techniques are applied for different biomass. Variations in the size of the produced chips are also possible because of the different equipment, thus resulting in different cost and consumption of energy (Hamelinck et al., 2005). This has implications as the physi- cal form of the wood biomass has effects on handling and transportation activities and costs. 122 Low density of the wood biofuels Wood chips produced are low density goods which require better capacity management in the vehicles and at the storage terminals. Better efficiency would lead to cost efficiency. Different densities of the biomass being transported results in different handling. As illustrated in figure 5.4.1, different forms of the wood biomass occupy different space but contain the same amount of biomass. The lower energy density of the wood biofuels makes transportation an important cost factor in the supply chains (Gold and Seuring, 2011). Since wood pellets and wood chips have gained the recognition of an internationally traded commodity, ship transport plays a vital role in the international trade of these commodities. However due to the low density of the wood pellets it presents a problem for the ship transport to be used. Ships, which are designed for ore transportations, cannot take up a full load of the pellets as it affects ship handling. With 7- 10 holds per ship, only 3-5 can be filled up with pellets and rest has to be filled with other heavy materials. So the materials left behind have to wait for the next ship to be transported. Specially designed ships for pellets do not have such problems, but then they offer limitations for other goods to be transported (Bradley et al., 2009a). Term standardisation There is no standard categorisation of the wood biofuels present in real situations. This reason makes it difficult to estimate the production and the consumption of the resources within the different regions. In that regard a standardised catalogue of the various wood biofuels would help in clear comparisons and trade of the wood biofuel entities. This also raises concerns about the already existing statistics of the wood biofuel industry (Frosch and Thorén, 2010). The categorisation of the wood bio fuels imported has been poor all over the world, which means the estimations of the production and consumption of the wood biofuels have been difficult in terms of how much is used for energy purposes. This unclear categorisation of the wood biofuels presents the problem of inaccurate reporting especially on the international scale of trade. It is further complicated by the fact that some indirect trade has also been done as round wood, which provides sawdust and bark as raw materials for pellet production. This Figure 5.4.1 Difference between dif- ferent biomass (Björheden et al., 2010). 123 categorisation problem is seen as a barrier for the international trade of the wood biofuels (Jungingera et al., 2008). Sources of supply Increase in the demand of the wood biofuels in the exporting countries can affect the imports of wood biofuels to Sweden. Imports can also be affected by the increase in the demand inter- nationally. As a result there would be more countries demanding the wood biofuels while the suppliers would be the same. In such scenarios prices of the wood biofuels will rise in the local markets along with reduced quantities (Hirsmark, 2002). Dependency on the policies In many EU countries, such as Sweden, the favourable policies for the generation of electrici- ty and other energies by the use of renewable sources have become the main drivers for the import of wood biofuels. However, some cases show that varying policies can result in in- creased prices of the biofuels while disturbing the market mechanism (Jungingera, Bolkesjøb et al., 2008). Swedish authorities have greatly shaped their policies for the propagation for the biofuels in the sectors of district heating. This is mainly due to fossil fuel taxation and subsid- iaries (Selkimaki et al., 2010). LOGISTICAL ANALYSIS The issues identified can be related to the supply chain management interfaces identified by Gold and Seuring (2011) in the bio energy supply chains. Table 6.1 provides a summary. Table 6.1 Logistical issues with references to SCM steps. chain steps. “X” denotes the pres- ence of the issue at various steps of SCM Logistical issues Operation 1 Operation 2 Operation 3 Operation 4 Overall sup- ply system design Difficulties Storage X Location, storage time, estimations, size, underutilisation of resources Seasonal variability X Keeping buffer stocks, uneven supply, increased storage costs Chipping process X X X X Uneven quality, location of the process, handling difficulties Low density of the wood biofuels X Handling complications, require capacity management both in vehi- cles and storage terminals Term standardisation X Difficulties in generating statistics Sources of supply X Shortage of wood biofuels Dependency on policies X Change in policies can lead to loss of interest in using wood biofuels 124 Seasonal variations of the wood biofuels require planning the stocks of wood biofuels to en- sure smooth production systems. Unstable supply can result in utilisation of fossil fuels in place of the renewable fuels to fulfil the demand, which is not environmentally friendly. The seasonal variations also mean that during the winters the demand is high while during the summers demand is almost zero. This results in the underutilisation of the resources at the plant such as machinery and equipment, which have fixed annual operation costs. The shorter the harvest periods, the more important storage becomes and its costs are increased. Mode of transportation utilised also varies due to the seasonal variability as larger volumes can utilise rail transportation during high demand in winters in comparison to summers when the demand is low. Storage terminals in the wood biofuel chain can be assumed to be the most essential part. It helps in managing the seasonal variability of the wood biofuels. There are a lot of decisions related to this integral part of the supply chain some of which are location, capacity and ac- cessibility to different modes. It’s also worth investigating which actors in the supply chain manage the storage terminals. The selection of a storage terminal depends upon the aforemen- tioned factors, which presents difficulties in storage capacity, storage time estimations and accessibility of the storage terminal. The seasonal variability of the wood biofuels may also result in the underutilisation of the resources at the storage terminals. The chipping process in the supply chain requires special attention as it affects the transporta- tion costs depending upon its position in the chain. The application of the chipping process makes wood biomass easier to transport and handle in comparison to whole trees. The posi- tion of the chipping process in the chain also depends upon the chipping equipment at a par- ticular site. The power plants that have chipping equipment available at their plants usually have good control over the quality of fuel that is fed to the burners whereas application of chipping previously in the chain helps in the efficient handling of the crushed wood biomass, thus ensuring maximum capacity utilisation of the vehicles. The low density of the wood biofuels also presents the problem of increased transport cost as the wood biofuel suppliers are paid based upon the energy content of the fuels they supply. Wood in the form of logging residues is heavy to transport with low energy content. In this regard the chipping process becomes really important in order to increase the energy content per volume of the wood biofuels supplied. The trucks carrying wood biomass are usually lim- ited in terms of volume in comparison to mass(Björheden et al., 2010). Usage of standardised terms is important for the reporting purposes in any business. Lack of standardised terms for the wood biofuels can lead to inaccurate reporting of which type of wood biofuel is needed and in what quantity. This issue becomes complicated when the consumers (CHPs and HPs) wouldn’t be able to determine what type and what quantity of wood biofuel they need for their power plants as no standardised terms makes it hard to know how much energy content a certain fuel has and how much quantity is required On a global level it can result in the hin- drance of trade due to the uncertainty of the terms being used for reporting and communica- tion between industries. Sweden receives large amounts of imported wood biofuels. The wood biofuels have met with a general success internationally. This will increase the global demand for wood biofuel and the risk of increased prices of the domestic supply of the wood biofuels The success of the wood biofuels in the energy sector was highly driven by European Union policies, including Swedish energy policies, which resulted in the increased use of renewable fuels instead of fossil fuels. Carbon tax in Sweden is one of the main drivers for the success of renewable fuels such as wood biofuels. These policies help in achieving the country’s energy efficient goals with subsidiaries and discouraging the use of fossil fuels. However, if the poli- 125 cies are to change drastically, it would become really difficult to compete with the energy and logistically efficient fossil fuels, which in the long term would increase the CO2 emissions of the whole country. Therefore the dependency of the wood biofuels on the policies makes them extremely volatile to sudden change in the policies. Consistency in the Swedish policies has greatly affected the success of biofuels, which included legislation, taxation, certificate sys- tems, fees and subsidies (Björheden et al., 2010). During the implementation of the wood bio- fuels in a region the biofuel chain can be subsidised, however in order to reach long-term sus- tainability large investments and commitments are required (Routa et al., 2012). Sweden has passed the implementation phase but policies such as the carbon taxes and certificate systems can be seen as valuable support for the sustainable use of wood biofuels. Immediate changes in the policies will result in a negative impact on the whole industry, including the developed logistical efforts for the distribution of wood biofuels. In the hope of finding more about how the process of transportation can be improved, one can call for an investigation of the proper application of the transport solutions. From the studies mentioned in this paper, development of long transportation networks of wood biofuels can be seen as both economically and environmentally feasible. The growing demand and environ- mental friendliness of wood biofuels makes them an interesting case to be studied in light of sustainable long distance transportation. The problems presented would provide valuable in- sights to developing sustainable transport solutions. CONCLUSIONS The study has shown that biofuel transport faces a number of important and unique challeng- es. Most importantly are the low density of the biofuels in combination with large volumes and uneven demand. The storage and chipping processes can be located at several different steps in the chain and are a key factor in the supply chain design. These issues are accompa- nied with the lack of standard terms of wood biofuels which would not only hinder trade but also result in inaccurate reporting. Finally, governmental policies have and will continue to play an important role in the success of wood biofuels. In the light of the identified issues, a call for sustainable transport solutions has been made, which is an interesting area for future research. Acknowledgements This research is funded by the Swedish Transport Administration (Trafikverket) through the project Sustainable intermodal supply systems for biofuel and bulk freight. REFERENCES: x AGENCY, T. S. E. 2012. Energy in Sweden - facts and figures 2012. The Swedish Energy Agency. x BJÖRHEDEN, R., ELIASSON, L., IWARSSON, M., HOFSTEN, H. V., ENSTRÖM, J. & JOHANNESON, T. 2010. Efficient forest fuel supply system. In: THORSEN, Å., BJÖRHEDEN, R. & ELIASSON, L. (eds.). Uppsala: Skogforsk. x BRADLEY, DIESENREITER, F., WILD, M. & TROMBORG, E. 2009a. World Biofuel Maritime Shipping Study. A study for IEA Bioenergy Task 40. 126 x BRADLEY, D., CUYPERS, D. & PELKMANS., L. 2009b. 2nd Generation Biofuels and Trade An exploratory study. IEA Bioenergy Task 40. x CARLSSON, D. & RÖNNQVIST, M. 2004. Supply chain management in forestry–– case studies at Södra Cell AB. European Journal of Operational Research, 163, 589- 616. x ERIKSSON, L. N. 2008. Comparative analyses of forest fuels in a life cycle perspective with a focus on transport systems. Resources Conservation and Recycling, 52, 1190-1197. x FROSCH, M. & THORÉN, P. 2010. Järnvägstransport av biobränslen. VÄRMEFORSK Service AB. x GOLD, S. & SEURING, S. 2011. Supply chain and logistics issues of bio-energy production. Journal of Cleaner Production, 19, 32-42. x GUNNARSSON, H., RONNQVIST, M. & LUNDGREN, J. T. 2004. Supply chain modelling of forest fuel. European Journal of Operational Research, 158, 103-123. x HAMELINCK, C. N., SUURS, R. A. A. & FAAIJ, A. P. C. 2005. International bioenergy transport costs and energy balance. Biomass & Bioenergy, 29, 114-134. x HIRSMARK, J. 2002. Densified Biomass Fuels in Sweden: Country report for the EU/INDEBIF project. Master of Science in Forestry, Swedish University of Agricultural Sciences. x JOHANSSON, J., LISS, J. E., GULLBERG, T. & BJÖRHEDEN, R. 2006. Transport and handling of forest energy bundles—advantages and problems. Biomass and Bioenergy, 30, 334–341. x JUNGINGERA, M., BOLKESJØB, T., BRADLEYC, D., DOLZAND, P., FAAIJA, A., HEINIMO, J., HEKTORF, B., LEISTADG, Ø., LINGH, E., PERRYI, M., PIACENTED, E., ROSILLO-CALLEI, F., RYCKMANSJ, Y., SCHOUWENBERGK, P.-P., SOLBERGL, B., TRØMBORGL, E., WALTERD, A. D. S. & WITA, M. D. 2008. Developments in international bioenergy trade. BIOMAS S AND B I OENERGY, 32, 717-729. x RAUCH, P. & GRONALT, M. 2010. The effects of rising energy costs and transportation mode mix on forest fuel procurement costs. Biomass and Bioenergy, 35, 690-699. x RENTIZELAS, A. A., TOLIS, A. J. & TATSIOPOULOS, I. P. 2008. Logistics issues of biomass: The storage problem and x the multi-biomass supply chain. Renewable and Sustainable Energy Reviews, 13, 887- 894. x ROUTA, J., ASIKAINEN, A., BJÖRHEDEN, R., LAITILA, J. & RÖSER, D. 2012. Forest energy procurement: state of the art in Finland and Sweden. WIRES Energy Environment. x SELKIMAKI, M., MOLA-YUDEGO, B., ROSER, D., PRINZ, R. & SIKANEN, L. 2010. Present and future trends in pellet markets, raw materials, and supply logistics in Sweden and Finland. Renewable & Sustainable Energy Reviews, 14, 3068-3075. x SVEBIO & ANDERSSON, K. 2012. Bionergy the Swedish Experience. Jönköping Sweden: Ark- Tryckaren. x TIMILSINA, G. R. & SHRESTHA, A. 2011. How much hope should we have for biofuels? Energy, 36, 2055-2069. 127 Paper 2 Awais, F. Flodén, J., 2013, Logistic requirements and characteristics of the Swedish wood biofuel industry. Submitted to the Scandinavian Journal of Forest Research. Logistic requirements and characteristics of the Swedish wood biofuel industry Awais, Fawad , & Flodén, Jonas Department of Business Administration, School of Business Economics and Law, University of Gothenburg, Sweden Postal address: Box 610, 405 30 Gothenburg, Sweden Phone: 031-786 2617, 031-786 5131 Abstract: Sweden has a large utilisation of forest fuels in the district heating plants (HPs) and combined heat and power plants (CHPs). This places large demands on the logistics system supplying these plants with fuel. The aim of this paper is to identify the industry actor’s re- quirements, constraints and preferences on the wood biofuel supply chain and to identify the logistical challenges this entails. To achieve this a survey was sent to all Swedish CHPs, com- bined with six interviews with transport companies, terminal operators and forest companies. The study shows that the industry has a local market focus, mostly utilising truck transport with direct transport from the forest. Road transport is rated highly favourable with reliability as the most important factor. Storage is used to overcome fluctuations in demand and is an essential part of the supply chain, with most CHPs have storage facilities. Comminution is most preferred in the forest. Challenges include determining the size and location of storages and identifying alternative possibilities for transport that might improve the transport chain and reduce environmental impact, while at the same time maintaining flexibility. Keywords: survey, interviews, logistical challenges, market, supply chain, transport, wood biofuels Acknowledgements: This work was supported by the Swedish Transport Administration (Trafikverket) under the project sustainable intermodal supply systems for biofuel and bulk freight. 128 Introduction The extent of economic and civil growth of mankind has been associated with the consump- tion of natural energy resources (Mikkilä et al., 2009). This growing demand for energy re- sources raises a number of problems. The use of fossil fuels has been the principal contributor to current environmental problems such as air pollution and greenhouse gas emissions. This calls for the development of renewable and environmental friendly energy resources. The use of biomass presents an interesting potential solution to these problems. Some of the many reasons for adoption of biomass as an energy source include its worldwide availability for use in power generation and the CO2 neutral basis of biofuels (Hamelinck et al., 2005). In the 1970s the interest of biofuels as an alternative to fossil fuel emerged due to the oil crisis (Timilsina and Shrestha, 2011, Björheden, 2006). This development was particularly promi- nent is Sweden (Björheden, 2006, Ericsson et al., 2004), that today has a large utilisation of forest fuels in the district heating plants (HPs) and combined heat and power plants (CHPs). District heating provides heating services to almost half of the Swedish population, with wood fuels generating 21 TWh for district heating in 2011 (Swedish Forest Agency, 2013), corre- sponding to 17 million solid cubic meters of wood fuels (Björheden et al., 2010). This large and increasing demand has placed increasing focus on the logistical issues of supplying the plants with fuel, as logistical issues are considered one of the key challenges for increased use of biofuel (Gold and Seuring, 2011, Svanberg and Halldórsson, 2013, Rentizelas et al., 2009). This increased demand and production of wood biofuels calls for a closer examination of the logistical activities involved in the transportation of these goods to the energy plants. An effective supply of biofuels faces a number of logistical challenges based upon costs, en- ergy use, material loss and applied logistics (Eriksson, 2008). Several studies have investigat- ed various aspects of the design of a biofuel supply chain (e.g., Svanberg et al., (2013); Gerasimov & Karjalainen (2013); Awais (2013); Oberscheider et al., (2012); Routa et al., (2012); Tahvanainen & Anttila (2011); Rauch & Gronalt (2010a, 2010b); Laitila (2008); Jo- hansson et al., (2006); Hamelinck et al., (2005); Gunnarsson et al., (2004)). The system is characterized by imbalances in the flows, seasonal variations and large transport volumes of low value goods with low density. Wood raw material is produced in the countryside, while the need for energy is greatest in cities. The need for energy is naturally greatest in winter and lower in summer. Harvesting sites within the forest also vary geographically. Awais (2013) identified seven key logistics challenges: storage, seasonal variability, chipping, low density, term standardisation, availability of biofuel and dependency on political policies. 129 However, these challenges cannot be studied in isolation. The design of any supply chain must take its starting point as the customer requirements it tries to meet, as a supply chain that does not meet the needs of the customer (in this case the HP/CHP) will certainly fail (Christo- pher, 2011, Bowersox, 2013, Simchi-Levi et al., 2008). This also includes understanding the constraints imposed by other actors in the supply chain. Any improvement in the biofuel sup- ply chain design therefore requires an understanding of the involved actor’s preferences, con- straints and requirements on the logistics system for biofuel transport and how these impact the logistics system. This study aims at investigating these issues in the Swedish market and to identify the logistical challenges involved. A typical supply chain This section will give a brief introduction to a typical supply chain for a CHP in Sweden. A more detailed description can be found in Routa, et al. (2012). A typical supply chain starts in the forest where the wood biofuels are harvested. Chipping (or other forms of comminution such as grinding) of the biomass commonly takes place here to increase the density of the biomass and improve transport efficiency (Björheden et al., 2010, Spinelli et al., 2011). Comminution is sometimes considered a part of harvesting (Gold and Seuring, 2011), but also commonly occurs at other location in the supply and can therefore also be considered a pre- treatment activity. The biomass is then transported by road, either directly to a power plant or to a terminal/storage. Road transportation is mainly used for the initial and final haul of the wood biomass and is performed by many types of trucks (switch body, side tilting, etc.) (Björheden et al., 2010). Rail or ship transport can be used for long distance transport from the terminal to the plant/other terminal. Another origin of the supply chain is the wood pro- cessing industry (e.g., saw mills), where by-products such as saw dust are collected. Storage is present at several locations such as at the power plants, terminals or along the road in the for- ests. Pre-treatment activities involve the chipping or comminution of the wood biomass and/or conversion to densified biofuels (e.g., pellets). Chipping is carried out either at the forests, terminals or the power plants, however, the densification is carried out at special production sites. Sometimes special trucks with mounted chippers are used in the forests (Eliasson and Picchi, 2010) while terminals and power plants have larger chipping equipment. 130 Gold and Seuring (2011) has divided the supply chain operations into five main parts. Table 1 Steps in the biofuel supply chain. Adapted from Gold and Seuring (2011). Category and key characteristics Actor’s involved Collecting/harvesting of the biofuel takes place in the forest. This may involve the chipping process, which affects the handling and costs of the wood biofuel transport. Forest owners Suppliers Storage helps when encountering seasonal variations for the demand and supply of wood biofuels, along with providing another option for the chipping process. Transport companies Terminal operators Power plants Transport is involved between the forests, storage terminals, heating plants, etc. The form of wood biofuels along with the transport modes (road, rail and sea) and load units/vehicles used are the key factors in this step. Transport companies Terminal operators Power plants Pre-treatment involves chipping, drying or conversion of wood biomass to densified biofuels. Pre-treatment has eco- nomic, environmental and social impacts on the supply chain. Forest owners Terminal operators Power plants Overall supply system design refers to the challenging task of effectively and efficiently designing and operating bio-energy production. Forest owners Transport companies Terminal operators Power plants 131 Figure 1 provides an overview of the flow of wood biomass and the various activities in- volved. Figure 1 National supply chains under focus. Triangles represent the onsite chipping and the dotted lines represent chipped biomass. 1* Operation 1: Harvesting and collecting biomass. 2* Operation 2: Storage. 3* Operation 3: Transport 4* Operation 4: Pre-treatment techniques. Materials and methods A web survey was distributed via e-mail during summer 2013 to a complete sample of man- agers at all 76 existing CHPs in Sweden using biofuels. Respondents not answering were phoned and urged to answer. The total response rate was 42% (n=32). The survey contained 132 30-38 questions and was made adaptive to the answers given. For example, if the responded did not use a certain type of fuel the survey adapted and removed related questions. The survey was divided into two parts. The first part of the survey included questions about the current equipment and practices at the power plants. These questions covered the general topics of storage, transportation, chipping and overall supply chain design. The second part investigated the perception of the power plants towards current and desired practices regard- ing the transport of wood biomass, which involved ranking the different practices or issues on a scale. The survey was supplemented by six telephone interviews and one e-mail interview to road (1), rail (1) and sea (e-mail) biofuel transport companies, a terminal company (1), forests company (1) and energy companies (2). The interviews lasted about 60 min each and were recorded. Results This section shows the results of the survey combined with results from the interviews. The total energy produced by the CHPs in the survey was 15.3 TWh (11.5 TWh winter October- March, 3.8TWh summer April-September) or 32% of the total energy production (48.1 TWh) by CHPs and HPs in Sweden (Svensk Fjärrvärme, 2010). The majority (75%) of this type of energy is produced during the winter. The respondents can be classified into three sizes based on winter energy production. Table 2 CHP size CHP sizes At the CHP At CHP and close-by Close-by No storage All 59% 28% 6% 6% Small 69% 25% 0% 6% Medium 80% 20% 0% 0% Large 0% 75% 25% 0% Fuel use is dominated by logging residue chips which account for ~25% of the energy gener- ated, followed by other wood chips at ~20%. Wood residue chips were the most widely used fuel with 75% of the respondents using this fuel, followed by chips of other wood (69%) and stem ships (50%). Essentially the same fuel mix is used in summer and winter. 133 Table 3 Fuel used Operation 1: Harvesting and collection Interview results show that the actors are not very interested in physical harvesting operations. Rather, they perceive that their interest start at the road side (road and forest energy actors), at the terminal (rail, terminal and shipping actors) or at the plant (energy actors). The rail, termi- nal, shipping and energy actor’s interest might extend further up the chain for some specific flows depending on contracts (e.g., if they are responsible for the transport from the forest), but actors are generally not interested in engaging in harvesting operations. However, they are well aware of the origin of their fuel and underline the importance of plan- ning ahead to secure a good supply of fuel. Contracts are commonly signed on a yearly basis for each winter season, and often the same supplier as the previous year is used. Contracts for several years also exist. Local suppliers are preferred due to transport costs and respondents perceive that there is a local monopoly/oligopoly, where the local CHP/HP(s) is often the only realistic buyer. Very little cooperation therefore exists between different plants on biofuel sourcing. Larger plants are forced to source from longer distances and thereby receive higher transport costs, called “quantity markup” by one respondent. The average number of suppliers for a CHP is approximately 14. The mean number of suppliers for the small power plants is 7 while the mean numbers of suppliers for medium and large power plants are 16 and 49 re- spectively. Most have one major supplier which stands for on average 44% of supply, with second and third suppliers accounting for an average of 21% and 14% of the CHPs’ energy requirement, respectively. Purchases are often made in and paid for according to energy con- Fuels Heat generated in winters Heat generated in summers Stem chips 9% 6% Logging residue chips 24% 22% Stub chips 2% 1% Chips of other wood 21% 25% Chips of unknown wood 9% 10% Pellets 2% 3% Peat 7% 3% Waste 8% 16% Bio gas 0% 0% Natural gas 0% 0% Bio oil 1% 0% Coal 1% 0% Oil 5% 0% Other fuels: 11% 14% Total 100% 100% 134 tent (MWh), but a large number of other units might be used depending on what is practically possible to measure and industry practice (e.g., fire wood is by tradition traded in m3 solid). Respondents perceive a good balance between supply and demand currently in Sweden. The resent increase in demand for biofuel of years steady is expected to level off as most major CHP/HPs in Sweden must soon be converted to biofuel plants or be biofuel new builds. How- ever, they see a risk of other industries influencing the fuel demand, for example potential increase in bioethanol production or economic fluctuations in the paper mill industry, which uses high quality wood chips for pulp. Export of biofuel is not considered feasible due to comparably high biofuel prices in Sweden and transport costs. A main obstacle is the limited development of district heating in Europe which keeps down the demand for biofuel and therefore also the fuel price. Import is used to a limited extent (1% of respondent’s energy). In general fuel prices are lower outside Sweden, which makes import economically possible, but the challenge is to keep transport costs down. Operation 2: Storage The vast majority (93.8%) of CHPs have storage options. Most (59%) store at the CHP only, 6% at a location close to the CHP and 29% at both. Only two respondents claim to have no storage at any location. Storage facilities at the CHP are smaller, with an average storage area of 60 GWh, than the close-by storage (often a terminal) that average 115 GWh. CHP storage is often limited by the physical space as most CHPs are located in residential areas. Storage times are also shorter at the CHP. Storage times are longer during the summer as consumption is less, however, fewer actors use storage during the summer. The use of close-by storage is greater for the large plants. Table 4 Share of respondents having storage option CHP sizes At the CHP At CHP and close-by Close-by No storage All 59% 28% 6% 6% Small 69% 25% 0% 6% Medium 80% 20% 0% 0% Large 0% 75% 25% 0% 135 Table 5 Average storage time in days, among respondents having storage Storage location Average storage time (days) winter Average storage time (days) summer CHP storage All 36 55 Small 50 63 Medium 26 49 Large 3 60 Close-by storage All 89 137 Small 113 114 Medium 137 183 Large 47 131 Storage is an issue for the CHPs, but is not considered problematic as most factors are rated of average importance. This is also supported by the interviews. More storage possibilities at the CHPs would be preferred but the current situation is accepted as extended storage capabilities are often physically impossible. Table 6 Mean ranking of storage problems (1=least problematic, 6=very problematic) Storage issues Total Small CHP Medium CPH Large CHP The location of storage facilities 3.7 3.4 3.5 4.8 The size of the storage facilities 3.8 3.8 3.5 4.8 The handling of biofuel at the storage facilities 3.4 3.2 3.3 4.0 The availability of equipment at storage facilities 3.2 2.9 3.4 3.5 Operation 3: Transport In winters, the Swedish forest is the origin for 51% of the energy while the forest industry accounts for 29%. Terminals are commonly used for the forest origin energy (25%) but less used for forest industry energy (6%). This includes both storage and transhipment terminals. Terminal shares decreases as the complexity of the supply chain increases with more interme- diate storage. Direct transport is most common (69% of respondents transport directly from forest while 59% of transport directly from forest industry in winters), as respondents state a desire to avoid terminals due to costs. The use of terminals is perceived to be increasing. 136 Table 7 Transport chains used Winter Summer CHPs utilising the chain Energy produced CHPs utilis- ing the chain Energy produced Directly from the forest 69% 26% 63% 39 From the forest via storage terminal 34% 12% 19% 3 From forest transhipment terminal 21% 7% 15% 8 From forest via both storage and tran- shipment terminal 7% 6% 7% 6 Directly from the forest industry such as saw mills 59% 23% 56% 21 From the forest industry via storage terminal 3% 0% 4% 0 From the forest industry via tranship- ment terminal 3% 0% 4% 0 From the forest industry via both stor- age and transhipment terminal 3% 6% 4% 6 From abroad 14% 1% 7% 1 Unknown chain 24% 19% 22% 16 Road is the most common mode of transport. All CHP get some deliveries by road, expect for one CHP that is supplied exclusively by conveyor belt from a nearby sawmill. Transport chains reveal that 72% of CHPs are completely supplied by road, corresponding to 84% of the total energy delivered (winter). The larger CHPs can receive 50-70 trucks per day. The supply chain design is essentially the same for winters and summers. Transport distances are also short where 47% of transport (counted in energy content) is shorter than 100 km. All trans- ports less than 250 km are performed by road only. CHPs try to keep the transport distance as short as possible due to the low value and density of the fuels. One respondent stated, “At 100 km it starts hurting and at 150 km it is really painful”. All CHPs and close-by storages have road access and most are equipped with a wheel loader (CHP: 91% Storage: 100%). Other common equipment include terminal area to position containers etc. (C: 34%, S: 64%), fixed cranes (C: 22%, S: 9%), trucks to handle detachable load units (C: 13%, S:18%) and possibil- ity switch body containers (C:59%, S:45%). Rail access is only present in 19% of the CHPs and used by one (3%). Ship access through quays is present at 13% of CHPs and is used by 9%. Access to rail and sea is limited to the medium (Rail:40%, Sea:20%) and large (Rail:50%, Sea:25%) CHPs. This is similar for close-by storage facilities, with 9% having rail access and 18% sea access. All CHPs that have such access use it. For both categories, rail is only used by large CHPs while sea is used by both medium and large CHPs. 137 Table 8 Transport distances R oa d on ly R ai l o nl y Sh ip o nl y T ra in , th en de liv er y by ro ad T ru ck , th en d el iv er y by tr ai n T ru ck , th en s hi p, d el iv - er ed b y tr uc k D el iv er y by t ru ck , pr ev i- ou s st ep s un kn ow n D is ta nc e, k m T ot al < 10 0 N = 23 10 0- 25 0 N = 15 25 0- 50 0 N = 3 > 50 0 N = 3 U nk no w n N = 3 T ot al U nk no w n T ot al 50 0- 75 0 T ot al N = 25 0- 50 0 N = 1 50 0- 75 0 N = 1 T ot al U nk no w n T ot al N = < 25 0 N = 1 U nk no w n T ot al < 25 0 N = 1 Percent of respondents 94 75 50 9 6 22 3 3 6 3 3 3 3 3 3 9 3 6 6 3 Percent of energy winter 84 47 17 3 3 14 2 2 1 1 4 2 2 5 5 2 1 1 2 2 As is often the case in logistics (e.g., Lammgård (2007); Saxin, et al., (2004) the most im- portant factor for the respondents is reliability. It is noteworthy that environmental sustaina- bility receives one of the lower rankings. Short transport time has been rated as the least im- portant factor, which is natural considering the low capital cost. Environmental aspects are given a low ranking, however, most CHPs are municipality owned and are thereby required to follow public purchasing regulations, which in many municipalities includes environmental standards. 138 Table 9 Ranking of important modal choice factors and service received (1=least ful- filled/important, 6=most fulfilled/important) Service Importance Service received Total Small CHP Medium CPH Large CHP Total Small CHP Medium CPH Large CHP Low cost 4.7 4.5 4.9 5.0 3.8 4.1 3.5 4.0 Short transport time 3.8 4.0 3.4 3.8 4.5 4.6 4.4 4.8 High reliability 5.1 5.1 4.9 5.5 4.7 4.9 4.3 4.5 High frequency 4.2 3.9 4.0 5.0 4.6 4.9 4.4 4.5 Few contaminations in the fuel, e.g. stones 4.9 4.8 4.5 6.0 4.1 4.4 3.6 4.0 Environmental sustain- ability 4.5 4.0 4.7 5.8 3.9 4.3 3.2 4.0 Good access to the transport system 4.7 4.3 4.9 5.8 4.7 4.7 4.6 4.8 For each transport mode, respondents were asked to state if the modes meet any of the seven qualities stated in Table 9 and the results are shown in Table 10. Trucks are rated most fa- vourable, as four of the seven factors received a rating above 50% of agreeing respondents. None of the other modes received any score above 50%. A comparison can be made with Ta- ble 9 showing that trucks match two of the top three criteria. A noteworthy exception is the environmental sustainability of trucks, which was not agreed to by any of the respondents. For this criterion, rail and sea had their highest scores at 48% and 39%. Trucks also scored low on low costs and contaminations in the fuel, however, this was given very low scores for all modes. For combinations of transport modes, a majority of the respondent did not agree to any of the qualities. The interviews identified the large volumes needed for viability and the inflexible system design as the main challenges to the use of rail and sea. Rail transport, in particular, has to be planned and scheduled months before the winter season, with limited possibilities for deviations. Interestingly, the larger plants that have more experience with rail and sea ranked these modes higher than the smaller plants. 139 Table 10 Qualities of different transport chains Qualities Total Small CHP Medium CHP Large CHP Trucks have low cost 16% 13% 10% 50% Trucks have short transport time 52% 50% 60% 50% Trucks have high reliability 64% 60% 70% 100% Trucks have high frequency 55% 50% 60% 75% Trucks have good access to the transport system, such as infrastruc- ture 68% 63% 80% 75% Trains are environmental sustaina- ble 48% 40% 60% 75% Trains have none of the specified characteristics 42% 50% 40% 0% Ships have none of the specified characteristics 45% 50% 40% 25% Combination of truck, train and ship have none of the specified characteristics 61% 63% 60% 25% Combination of trucks and trains have none of the specified charac- teristics 58% 63% 40% 25% Combination of trucks and ships have none of the specified charac- teristics 52% 13% 10% 50% The preference for truck transport is supported by the ranking of preferred transport modes, showing a very clear advantage for road transport. Table 11 Mean ranking preferred transport mode (1=least preferred, 6=most preferred) Mode of transport Ranking Road 5.5 Rail 2.5 Truck and ship combined 2.5 Ship 2.3 Truck and rail combined 2.3 Truck, rail and ship combined 2.2 Transport of different wood biofuels is considered more problematic by the larger CHPs with larger flows. Tree parts and forest residues are the most problematic fuels. The ranking corre- sponds well to the density of the fuels, which itself was ranked as a top problem. 140 Table 12 Mean ranking of transport problems (1=least problematic, 6=very problematic) Total Small CHP Medium CHP Large CHP Transport of forest residues 3.0 2.8 3.3 3.5 Transport of tree parts 3.1 2.9 3.4 3.5 Transport of wood chips 2.5 2.5 2.7 2.5 Transport of pellets 2.3 2.1 2.5 3.0 The availability of suitable transport 3.0 2.7 4.0 2.3 The low density of biofuel 3.6 3.3 3.7 4.0 Risk of contamination of the biofuel during transport, e.g. stones 3.5 3.6 3.3 3.3 Most CHPs allow their suppliers to organise the transport and purchase the fuel with transport included. Most (48%) CHPs have their transports completely arranged by suppliers. No dif- ference can be seen depending on the size of the plant. Interviews showed that the CHPs mostly arranged transports from the CHPs own close-by storage and from the forest industry to their plant. External transport providers were contracted. The mode of transport was decided by the CHP for 54% of the CHPs, by the supplier for 48% and by the transport company for 13%. This is supported by the interviews that showed that the CHPs retain influence over the transport and would not allow any transport system against their will. Only one respondent in the survey did not know the transport chain used, which is low compared to other industries. Rail and sea transport has been considered for use by 37% and 41% of CHPs respectively. The costs levels for all modes of transport are mainly un- known by the power plants. The supplier pays for 75% of the total energy transported while plants pay for 23% of energy transported. On average 89% of CHPs transports are paid for by the supplier, while an average of 50% of the CHPs have all transports paid for by the supplier. The most common load units/vehicles are the fixed/tilting truck, which are used to some ex- tent by 88% of the CHPs and for 67% of the energy, followed by switch body trucks by 75% of the CHPs and 25% of the energy. Other vehicles used by the CHPs include timber trucks and conveyor belts. Interviews revealed that fixed/tilting trucks are more often used from ter- minals/storage/industry and switch body systems more common when picking up in the for- est, due to that the switch body trucks are more accessible and can drop-off load units in the forest, but have a lower loading capacity. For trains, the Innofreight container is the only load unit used. 141 Table 13 Load units/vehicles used Load units for trucks Percentage of energy produced Percentage of usage Small CHP Medium CHP Large CHP Truck with fixed / tilting superstruc- ture, e.g. side -dump truck 67% 75% 100% 100% Container truck with switch body, e.g. wood chip container 25% 69% 90% 100% Other 6% 19% 0% 25% Do not know 2% 6% 0% 0% Trucks have the shortest unloading time with an average of 27 minutes at the plants and 30 minutes at the close-by storage. However, estimates range from just 5 to 60 minutes. The time includes all activities from when the vehicle arrives at the gates, (e.g., reception, any meas- urements, moving of load units etc.). Rail is unloaded at 4 hours or 3.6 minutes per load units (66 per train) at both plants and terminals. Unloading of ships shows a greater variety with 15- 48 hours, with ship sizes ranging from 2000 to 4000 tonnes. Interestingly, the unloading time does not seem to correspond to the ship size. For example, a 3000 tonne ship is reported to be unloaded in 15 hours, while a 2000 tonne ship is unloaded in 23 hours. This is supported by the interviews that identify unloading of the ships as a troublesome area. Most ports are con- sidered not customer oriented and to a large extent controlled by inflexible labour union rules. Opening hours to receive fuels are flexible with most CHPs open weekdays 06.00- 22.00. Some are open 24 hours per day, 7 days per week. Most receive their fuel during daytime. Operation 4: Pre-treatment techniques Chipping of wood biofuels is mostly performed in the forest and is also the most preferred option. Often, chipping is considered a part of the harvesting (Gold and Seuring, 2011), but can also occur at other locations in the chain as a pre-treatment activity. However, chipping at the CHP is preferred by the large CHPs due to a perceived better con- trol over the fuels, lower chipping cost, better quality and higher efficiency. Environmental aspects are also mentioned since a terminal chipper can be powered by electricity. Obstacles stated are local laws prohibiting chipping at CHPs in urban areas due to noise and dust, and that high volumes are need to be economically feasible. This is also shown by the fact that CHP chipping is most common among the larger CHPs. Smaller CHPs prefer receiving the fuel already chipped. Most CHPs know where the chipping is performed, however, unknown locations represent a surprisingly large share of the energy. The majority of this can be traced back to a single very large CHP with a high share unknown. 142 Table 14 Chipping location used in energy of chipped wood and preference (1=least pre- ferred, 6=most preferred) Total Small CHP Medium CHP Large CHP Chipping location C H P s us in g Sh ar e of en er gy P re fe re nc e C H P s us in g P re fe re nc e C H P s us in g P re fe re nc e C H P s us in g P re fe re nc e Chipped at the forest 66% 42% 4.6 53% 4.4 80% 5.1 75% 3.8 Chipped at the terminal 55% 18% 4.1 47% 4.6 60% 3.8 50% 3.0 Chipped at the CHP 41% 19% 3.4 27% 2.8 50% 3.6 75% 5.0 Chipping location unknown 41% 19% - 53% - 20% - 25% - Overall operation of the supply chain Variations in demand (e.g., due to low outside temperature requiring more district heating) are mainly dealt with by buffer stocks and extra deliveries. Interviews show that many CHPs con- sider this to be their supplier’s problem. Contracts commonly state an annual volume to be delivered with a request made the week before (normally on Thursday) of what is needed the following week. Volumes are expected to be flexible and often defined within ranges (+/- X%) in the contract. Changes seldom cause any conflicts, although respondents do not think the systems works well and call it a “guessing game”. Better long term weather forecasts are desired. Table 15 Handling fluctuations in demand. Respondents using the option. Options used Total Buffer stocks 74% Extra deliveries 71% Energy storage 35% Use of fossil fuels 29% Cooperation with other heating plants 13% The fluctuating demands of the biofuels are handled similarly by the different categories of the plants. Cooperation with other plants and energy storage are more common among the larger plants, while smaller plants rely more on buffer stocks and deliveries. The most important factors in the biofuel supply chain are on time deliveries, fuel quality and no contamination in the fuel. Comparing the preferences and services received reveals that the CHPs are in general satisfied with their supply chain. However, suppliers only meet their full expectations on the flexible delivery option. On all other factors, the service received is less 143 than their importance. The low scores given to low cost of transport and biofuel, in compari- son to their high importance, is particularly noteworthy. Table 16 Service received and their importance (1=least fulfilled/important, 6=most ful- filled/important) Service Service received Importance Total Small CHP Medium CPH Large CHP Total Small CHP Medium CPH Large CHP Flexibility concerning delivery options 4.3 4.3 4.3 4.3 4.3 4.3 4.1 5.0 Flexibility concerning ordered volumes 4.4 4.8 4.1 4.3 5.0 4.9 5.1 5.3 On-time deliveries 4.4 4.8 4.4 3.8 5.4 5.4 5.2 5.8 Low cost transport 3.6 3.9 3.4 3.3 4.8 4.6 4.9 5.0 Low biofuel price 3.5 3.9 3.2 2.8 5.4 5.4 5.7 5.0 Small environmental impact 4.2 4.3 4.1 4.3 5.0 4.8 5.2 5.5 Good quality biofuel 4.5 4.5 4.5 4.3 5.5 5.6 5.4 5.8 No contamination in the delivered fuel, e.g., no stones. 4.1 4.1 4.1 3.5 5.6 5.6 6.0 6.0 Deliveries evenly dis- tributed in time 3.9 4.0 4.0 3.0 4.8 4.7 5.0 5.0 On a more general level, the largest problem perceived was the dependency on Swedish polit- ical decisions. In the interviews, a clear distinction could be made between CHPs and other actors, where the other actors did not perceive to be dependent on politics. CHPs are often municipality owned and subject to emission taxes and other legislation. At the same time, the interviewees perceived Sweden to be a world leader in biofuels for CHP, due to the early po- litical decision to build extensive district heating which created a market. In general, the mar- ket appears well functioning with no major cooperation problems or lack of fuel. Table 17 Issues in the Swedish biofuel industry (1=least problematic, 6=most problematic) Issue Mean rating Dependence on Swedish political decisions on biofuels 4.0 Seasonal variations in the demand for heating 4.3 Contamination of the biofuel at delivery, e.g. stones 3.8 Dependence on foreign policy decisions on biofuels 4.0 The biodegradation of the biofuel 3.9 The drying of biofuel 3.3 Standardized terms and definitions for bio-fuels and raw materials 2.7 Impact on the Swedish biofuel market from increased global demand for biofuels 3.1 The availability of biomass 2.6 Cooperation between actors in the industry 3.0 144 Analysis The share of bioenergy around the world is considered to increase in the coming decades. A clear understanding of the biofuel market operations would facilitate the decision makers for making valuable choices (Roos et al., 2000). The actor’s preferences and current supply chain characteristics, as described in the results chapter, can be used to highlight important chal- lenges for the logistics system. Operation 1: Harvesting and collection The local focus in the supply chain is clearly visible. Truck transport is only economically viable for shorter distances and other transport modes require large transport volumes. In this way the market is limited for smaller CHPs, who are essentially restricted to source within 100-150 km of the plant by truck. Larger plants have more sourcing options, however, also prefer local sourcing if possible. Lack of local fuel is the main driver for long distance sourc- ing, as transport costs seldom make it possible for long distance fuel to match local prices. In Sweden, the competition for biofuel is most intense in the densely populated central-eastern regions (Roos et al., 2000). Little cooperation exists on biofuel sourcing among the CHPs, although studies have shown that a co-operative strategy could reduce transport costs (Rauch et al., 2010). In contrast to many other industries, a CHP does not have the option of moving to a more favourable sourcing location, as it must be connected to the local district heating grid. A clear difference can be seen in the procurement between products directly from the forest and forest industry residues, such as saw dust from saw mills. The forest chain is a pull-chain, where the CHPs order the amounts they need. The industry residues are a push-chain, where the industry production sets the pace of which the by-products are produced. Contracts often state that the CHP agrees to accept all by-products produced, as the industry sees it as a waste product to get rid of. The power in the chains are different where the CHP is the channel lead- er in the forest chain and the supply chain is set according to the CHPs need, while the forest industry is the leader in the industry chain and the chain is set according to the industry needs. Operation 2: Storage Storage is a vital part of the wood biofuel supply chains and the results demonstrate that it is widely used by the CHPs. The uneven demand of the district heating plants has created the need of storage. The demands are highest in the season of autumn and winter, therefore wood biofuels are stored for at least a few months. In general the moisture content of the wood bio- 145 mass decreases with the increase of storage time. Sometimes the moisture content rises in autumn or winters which again requires the biomass to be kept in storage for even longer pe- riods of time (Anerud and Jirjis, 2011). From a logistical point of view, the use of storage enables more efficient transport flows and is a requirement in high volumes flows (e.g., ships and trains) supplying more than the immediate fuel need. Storage also makes it possible to handle seasonal variations. The wide spread use of storage gives greater flexibility in design- ing the supply chain, at the same time as it contributes to a significant share of the cost. As all storage costs money, there is a challenge in balancing the size of storage against the costs. The biofuel industry is further constrained by the limited physical area at many plants, which has caused CHPs to establish close-by storage areas. Dividing the storage causes higher logis- tical costs, which must be outweighed by the gain from having more storage area. Rentizelas et al. (2009) concludes that cheap storage solutions can help in reducing the overall costs of the biofuel supply chains. This reduced cost can even compensate the material losses and in- creased handling at the storage terminals. Operation 3: Transport Transportation of wood biofuels consumes the most fossil fuels in the wood biofuel supply chains. Previous studies indicate that the employment of rail and road would require a lower energy requirement for transportation, than relying on road only. Trains can carry much big- ger volumes than trucks and although being more efficient this also require transport distances to be long (Lindholm and Berg, 2005).The CHPs prefer reliable transport, with good access, low costs and few contaminations in the fuel, and perceive that they receive this from road transport. This in combination with the short transport distances often makes road the only realistic transport option. However, to use trucks only to satisfy the demand of a large CHP can generate traffic congestion problems due to the large number of trucks needed (Mahmudi and Flynn, 2006). The survey showed that large plants can receive up to 70 trucks per day. Shifting the transportation load to other modes is one alternative solution. However, other transport modes have problems matching the requirements, with the exception of environmental sustainability, where sea and rail are rated high and road is rated low. A major challenge for sea and rail is the lack of infrastructure access to the CHPs. Ships, which are designed for ore transportations, cannot take up a full load of the pellets as it affects ship handling. With 7-10 holds per ship, only 3-5 can be filled with pellets and rest must be filled with other heavy materials. Therefore the materials left behind have to wait for the next ship 146 to be transported. Specially designed ships for pellets do not have such problems, but offer limitations for other goods to be transported (Bradley et al., 2009) This can be overcome by intermodal transport, where several transport modes are combined (UN/ECE, 2001), for example, a rail transport to a terminal followed by a road transport to the CHP. The biofuel supply chain can be seen to have potential for intermodal transport due to the wide use of storage and transhipment terminals which can serve as intermediate points in the chain. However, intermodal transport requires large volumes and long transport distances to be competitive, as rail/sea has low distance dependent transport costs but high fixed costs (Flodén, 2007). Extra terminal costs are also added, unless the fuels should have passed a terminal anyway. Long distance intermodal transport is slower than local road transport, but fast transport was ranked of low importance. A key actor in developing alternative transport solutions is the supplier, as they are often responsible for the arranging the transport. Howev- er, the CHP holds a large influence over the transport chain and any radically new transport solution would require the approval of the CHP. The varying fuel demands of CHPs pose logistical problems as the transport equipment utili- sation will vary from one week to another, causing unnecessary cost for idle machinery. Out- side use of the equipment will help reduce the costs. It can be noted that the transport preferences shown by the CHPs corresponds well to the preferences found in other industries, where transport quality and reliability is ranked of high- est importance. The transport choice is described as a two-step process of first determining if the key preferences are met, after which the transport options are evaluated on price (Flodén et al., 2010). Operation 4: Pre-treatment techniques The productivity of the chipping equipment is high at the terminal in comparison to the forest because there is no set up time along with predefined space. Chipping at a terminal gives con- trol over quality, however, waiting times between the chipper and the truck would reduce productivity. Terminal chipping also requires large volumes to stay economically competitive. Delays due to queuing or transport distances result in increased costs (Asikainen, 1998). The location of the chipping process has a large impact on the logistics as it increases the density of the fuel and thereby the possible transport options and number of vehicles needed. The closer the forest in which the chipping takes place, the more efficient the transport utilisation can be (except for round wood). This is balanced against the more cost efficient chipping with higher quality that can be made at terminals and CHPs where the higher volumes makes it 147 possible to use more efficient equipment. Most chipping currently takes place in the forest, which suggests that the current balance of efficiency lies here, which is also the preferred lo- cation. The preference for chipping at the CHP among the large CHPs poses a logistical chal- lenge as this reduces the transport utilisation. Overall operation of the supply chain Biofuels are used extensively due to the high carbon tax as the industry has implemented measures to reduce the use of fossil fuels (Svensk Fjärrvärme, 2010). The supply chains are characterised by fluctuating demand and are highly dependent on buffer stocks and extra de- liveries. The supply chain must be flexible to adapt to these changes. Road transport this has the advantage of being a very flexible transport mode, in comparison with sea and rail that operate on longer planning horizons. An outside threat is the influence of political decisions and policies, which on short notice, can change the operational conditions. A possible criti- cism of the biofuel friendly policies is that such initiatives require investments, which have to be supported, by regulation, taxation and subsidization. The money invested can be used to advance the economic growth of a country, which would result in a stronger economy by pos- ing as income of the people (Walter, 2012). Conclusions The following table provides a summary of the logistical challenges identified: Table 28 Summary of the logistical challenges Operations Current situation Logistical challenges Harvesting and collection Local market focus with truck transport. Local focus limits the possi- ble logistical and sourcing solutions. Storage Essential in the supply chain at both the plants and terminals. Balancing the size and loca- tion against the extra costs. Transport Highly dominated by road. The supply chain structure is mainly transport directly from the forests. Finding possibilities for other transport modes to potentially improve the chain, e.g. inter- modal transport. Reducing environmental impact from road transport. Pre- treatment Most common and preferred loca- tions are the forests and the termi- nals. Balancing between more effi- cient chipping and more effi- cient transport. 148 Overall supply chain Fluctuating demand and buffer stocks. Influenced by political decision. Flexibility in the supply chain to adapt to changes. The results presented in the survey and the interview study provides information on the cur- rent situation of the industry along with the possible logistical challenges that are present in the wood biofuel supply chain. Literature studies e.g., Awais (2013) have identified storage, seasonal variability, chipping, low density, term standardisation, availability of biofuel and dependency on policies as logistical issues for biofuel transport. This study can confirm most of these issues, with the exceptions of standardisation and availability of fuel, which were not considered a problem by the respondents. References 1. ANERUD, E. & JIRJIS, R. 2011. Fuel quality of Norway spruce stumps – influence of harvesting technique and storage method. Scandinavian Journal of Forest Re- search, 26, 257-266. 2. ASIKAINEN, A. 1998. Chipping terminal logistics. Scandinavian Journal of Forest Research, 13, 386-392. 3. AWAIS, F. 2013. Wood biofuels logistical challenges in Sweden. NOFOMA 2013. Gothenburg, Sweden. 4. BJÖRHEDEN, R. 2006. Drivers behind the development of forest energy in Sweden. Biomass and Bioenergy, 30, 289-295. 5. BJÖRHEDEN, R., ELIASSON, L., IWARSSON, M., VON HOFSTEN, H., EN- STRÖM, J. & JOHANNESON, T. 2010. Efficient forest fuel supply system. In: THORSEN, Å., BJÖRHEDEN, R. & ELIASSON, L. (eds.). Uppsala: Skogforsk. 6. BOWERSOX, D. J. 2013. Supply chain logistics management, New York, McGraw- Hill. 7. BRADLEY, DIESENREITER, F., WILD, M. & TROMBORG, E. 2009. World Bio- fuel Maritime Shipping Study. A study for IEA Bioenergy Task 40. 8. CHRISTOPHER, M. 2011. Logistics and supply chain management : creating value- adding networks, Harlow, England ; New York, Financial Times Prentice Hall. 9. ELIASSON, L. & PICCHI, G. 2010. Huggbilar med lastväxlare och containrar. Ar- betsrapport. Uppsala: Skogforsk The Forestry Research Institute of Sweden. 10. ERICSSON, K., HUTTUNEN, S., NILSSON, L. J. & SVENNINGSSON, P. 2004. Bioenergy policy and market development in Finland and Sweden. Energy policy, 32, 1707-1721. 11. ERIKSSON, L. N. 2008. Comparative analyses of forest fuels in a life cycle perspec- tive with a focus on transport systems. Resources Conservation and Recycling, 52, 1190-1197. 12. FLODÉN, J. 2007. Modelling intermodal freight transport : the potential of com- bined transport in Sweden. Doctoral thesis, School of Business, Economics and Law at University of Gothenburg, BAS Publishing. 149 13. FLODÉN, J., BÄRTHEL, F. & SORKINA, E. Factors influencing transport buyers choice of transport service - A European literature review 12th WCTR conference, July 11-15 2010 Lisbon. 14. GERASIMOV, Y. & KARJALAINEN, T. 2013. Energy wood resources availability and delivery cost in Northwest Russia. Scandinavian Journal of Forest Research, 28, 689-700. 15. GOLD, S. & SEURING, S. 2011. Supply chain and logistics issues of bio-energy production. Journal of Cleaner Production, 19, 32-42. 16. GUNNARSSON, H., RONNQVIST, M. & LUNDGREN, J. T. 2004. Supply chain modelling of forest fuel. European Journal of Operational Research, 158, 103-123. 17. HAMELINCK, C. N., SUURS, R. A. A. & FAAIJ, A. P. C. 2005. International bio- energy transport costs and energy balance. Biomass & Bioenergy, 29, 114-134. 18. JOHANSSON, J., LISS, J.-E., GULLBERG, T. & BJÖRHEDEN, R. 2006. Transport and handling of forest energy bundles—advantages and problems. Biomass and Bio- energy, 30, 334–341. 19. LAITILA, J. 2008. Harvesting Technology and the Cost of Fuel Chips from Early Thinnings. Silva Fennica, 42, 267–283. 20. LAMMGÅRD, C. 2007. Environmental marketing of freight transports - the inter- modal road-rail case. Doctoral thesis, School of Business, Economics and Law at University of Gothenburg. 21. LINDHOLM, E. L. & BERG, S. 2005. Energy requirement and environmental im- pact in timber transport. Scandinavian Journal of Forest Research, 20, 184-191. 22. MAHMUDI, H. & FLYNN, P. C. 2006. Rail vs Truck transport for biomass. Applied biochemistry and biotechnology, 129-132. 23. MIKKILÄ, M., HEINIMÖ, J., PANAPANAAN, V., LINNANEN, L. & FAAIJ, A. 2009. Evaluation of Sustainability schemes for international bioenergy flows. Inter- national journal of energy sector management, 3, 359-382. 24. OBERSCHEIDER, M., ZAZGORNIK, J., HENRIKSEN, C. B., GRONALT, M. & HIRSCH, P. 2012. Minimizing driving times and greenhouse gas emissions in timber transport with a near-exact solution approach. Scandinavian Journal of Forest Re- search, 28, 493-506. 25. RAUCH, P. & GRONALT, M. 2010a. The effects of rising energy costs and trans- portation mode mix on forest fuel procurement costs. Biomass and Bioenergy, 35, 690-699. 26. RAUCH, P. & GRONALT, M. 2010b. The Terminal Location Problem in the Forest Fuels Supply Network. International Journal of Forest Engineering, 21, 32-40. 27. RAUCH, P., GRONALT, M. & HIRSCH, P. 2010. Co-operative forest fuel pro- curement strategy and its saving effects on overall transportation costs. Scandinavian Journal of Forest Research, 25, 251-261. 28. RENTIZELAS, A. A., TOLIS, A. J. & TATSIOPOULOS, I. P. 2009. Logistics is- sues of biomass: The storage problem and the multi-biomass supply chain. Renewa- ble & Sustainable Energy Reviews, 13, 887-894. 29. ROOS, A., BOHLIN, F., HEKTOR, B., HILLRING, B. & PARIKKA, M. 2000. A geographical analysis of the Swedish woodfuel market. Scandinavian Journal of For- est Research, 15, 112-118. 30. ROUTA, J., ASIKAINEN, A., BJÖRHEDEN, R., LAITILA, J. & RÖSER, D. 2012. Forest energy procurement: state of the art in Finland and Sweden. WIRES Energy Environment, . 150 31. SAXIN, B., LAMMGÅRD, C. & FLODÉN, J. Meeting the demand of goods transport - identification of needs among Swedish companies. NOFOMA 2004, 7-8 june 2004 Linköping, Sweden. 32. SIMCHI-LEVI, D., KAMINSKY, P. & SIMCHI-LEVI, E. 2008. Designing and managing the supply chain : concepts, strategies, and case studies, Boston, McGraw- Hill/Irwin,. 33. SPINELLI, R., CAVALLO, E., FACELLO, A., MAGAGNOTTI, N., NATI, C. & PALETTO, G. 2011. Performance and energy efficiency of alternative comminution principles: Chipping versus grinding. Scandinavian Journal of Forest Research, 27, 393-400. 34. SVANBERG, M. & HALLDÓRSSON, Á. 2013. Supply chain configuration for bi- omass-to-energy: the case of torrefaction. International Journal of Energy Sector Management, 7, 65-83. 35. SVANBERG, M., OLOFSSON, I., FLODÉN, J. & NORDIN, A. 2013. Analysing biomass torrefaction supply chain costs. Bioresource Technology, 142, 287-296. 36. SVENSK FJÄRRVÄRME. 2010. The Swedish District Heating Association Homep- age [Online]. Available: http://www.svenskfjarrvarme.se/Statistik-- Pris/Fjarrvarme/Energitillforsel/ [Accessed 2012 January 6. 37. SWEDISH FOREST AGENCY 2013. Swedish Statistical Yearbook of Forestry. Jönköping: Swedish Forest Agency, Skogsstyrelsen. 38. TAHVANAINEN, T. & ANTTILA, P. 2011. Supply chain cost analysis of long- distance transportation of energy wood in Finland. Biomass and Bioenergy, 35, 3360-3375. 39. TIMILSINA, G. R. & SHRESTHA, A. 2011. How much hope should we have for biofuels? Energy, 36, 2055-2069. 40. UN/ECE 2001. Terminology on Combined Transport. New York and Geneva: Unit- ed Nations (UN) & Economic Commission for Europe (ECE). 41. WALTER, S. 2012. The biofuel bubble in northern Sweden and Finland. Internation- al Journal of Energy Sector Management, 6. 151 Paper 3 Flodén, J., Awais, F., 2014. Meeting the challenges for intermodal transportation of biofuel. Meeting the challenges for intermodal transportation of biofuel Flodén, Jonas Awais, Fawad Department of Business Administration, School of Business Economics and Law University of Gothenburg Sweden Postal address: Box 610, 405 30 Gothenburg, Sweden Abstract: The use of solid biofuels for energy in heating plants has increased drastically during the last decades. This substantial and increasing demand has placed focus on delivering the supply to the plants, as logistics issues are considered one of the key challenges for further increased use of biofuel. Environmental concerns, increasing size of power plants, and challenges in sourcing enough fuel locally has sparked an interest in using intermodal road-rail transport. A case study is formed at a Swedish district heating plant to investigate the potential of introduc- ing intermodal transport. Extensive calculations are performed into the design and operations of an intermodal system, showing both costs and CO2 emissions. This is analysed in relation to key logistical challenges in the industry. A best feasible case scenario is identified. Conclu- sions are that the potential for intermodal transport is greatest among the largest plants with large volumes to achieve high resource utilisation. An advantage for intermodal transport is that large flows currently pass through a terminal, which improves the competitiveness against road transport and allows for use of efficient resources at the terminal. This study leads to better understanding of the strengths and weaknesses of intermodal biofuel transport and has practical implications for anyone in the process of designing such systems. Keywords: biofuel, road, rail, intermodal, CHP, wood chips, transport, case 152 Introduction The use of solid biofuels for energy in heating plants increased drastically during the decades after the oil crisis in the 1970s (Timilsina and Shrestha, 2011, Björheden, 2006). This devel- opment was particularly prominent in Sweden (Björheden, 2006, Ericsson et al., 2004), which today largely utilises forest fuels in district heating plants (HP) and combined heat and power plants (CHP). From practically nothing in 1970, wood biofuels today account for 21 TWh, or 35%, of the total energy used for district heating in Sweden (Energimyndigheten, 2012, Swedish Forest Agency, 2013) . This substantial and increasing demand has placed focus on delivering the supply to the plants, as logistics issues are considered one of the key challenges for further increased use of biofuel (Gold and Seuring, 2011, Svanberg and Halldórsson, 2013, Rentizelas et al., 2009). Keeping logistics costs low is important for a competitive bio- fuel system, which is of key importance in reducing both dependence on fossil fuels and the greenhouse effect. Savings in greenhouse gases amount to approximately 90% when fossil fuel is replaced with biofuel. Biofuel also brings a very positive energy balance, as the energy consumed to produce biofuel only represents 2-5% of the energy in the biofuel, compared to 15% for fossil fuel (Lindholm, 2010). Logistical challenges An effective supply of biofuels faces a number of logistical challenges based upon costs, en- ergy use, material loss, and applied logistics (Eriksson, 2008). The system is characterised by imbalances in the flows, seasonal variations, and large transport volumes of low value goods with low density. Wood raw material is produced in the countryside, while the need for ener- gy is greatest in cities. The need for energy is greatest during winter and lowest in summer. Harvesting sites within the forest also vary geographically. Awais and Flodén (2014) identified five key logistical challenges: seasonal variability, stor- age, chipping, low density, and dependency on political policies. In wood biofuel chains the stocks vary a great degree depending on seasonal demand and seasonal forest management. This results in the underutilisation of expensive machinery and equipment, causing an in- crease in annual operational costs. Due to the climate, the need for energy is greatest during cold winters and very low during warm summers. Most biofuel HPs close down during the summer and only operate between approximately late August and April. The daily demand also varies significantly due to the outside temperature. In Sweden, the traditional harvesting period of wood biofuels is during autumn, winter, and early spring. Buffer stocks are therefore needed for a smooth flow of raw materials and to have enough stocked for the months of Au- gust and early September, after which the harvesting processes begin again (Carlsson and Rönnqvist, 2004). Chipping, or reducing the wood to smaller chips, is also a part of the supply chain. This increases the density of most biofuels, allowing for more efficient transport. Chip- ping is normally performed at the roadside in the forest, but can also be performed at the stor- age terminal or at the HP. Different machines are commonly used at the different locations, thus resulting in different costs and energy consumption (Hamelinck et al., 2005). Chipping at a terminal/HP is more efficient but requires a low density transport from the forest. In general, wood biofuel, both chipped and unchipped, is a low density good that requires good capacity management in the vehicles and at the storage terminals. This low density makes transport an important cost factor in the supply chains (Gold and Seuring, 2011). Political interest in bio- fuels is great as a means to reach a more sustainable society and to reduce CO2 emissions and climate change. Most district heating plants are municipality owned (Svensk Fjärrvärme, 2014) and are thereby subject to political decisions, such as environmental requirements in 153 tendering. The logistics system is also subject to legal and political regulations on vehicle sizes, emission levels, fuel taxes, permits, emission trading, etc. The supply chain challenges can be further explained according to the five main parts of the supply chain: Harvesting and collection, Storage, Transport, Pre-treatment, and Overall man- agement. Table 4 Summary of the logistical challenges (Awais and Flodén, 2014) Operations Current situation Challenges Harvesting and col- lection Local market focus with truck transport. Local focus limits the possible logistical and sourcing solutions. Storage Essential in the supply chain at both the plants and terminals. Balancing the size and location against the extra costs. Transport Highly dominated by road. The supply chain structure is mainly transported directly from the forests. Finding possibilities for other traf- fic modes to potentially improve the chain; e.g., intermodal transport. Reducing environmental impact from road transport. Pre- treatment Most common and preferred locations for chipping are the forests and the ter- minals. Balancing between more efficient chipping and more efficient transport. Overall supply chain Fluctuating demand and buffer stocks. Influenced by political decisions. Flexibility in the supply chain to adapt to changes. Problem description Currently, 84% of the energy from wood biofuels to Swedish CHPs is transported by road only and road is the traffic mode the industry perceives to best meet their current logistics requirements (Awais and Flodén, 2014). However, environmental concerns, the increasing size of power plants, and challenges in sourcing enough fuel locally has sparked an interest in other traffic modes, such as rail and also partly by sea (Enström and Winberg, 2009, Frosch and Thorén, 2010, Mahmudi and Flynn, 2006, Björheden et al., 2010, Tahvanainen and Anttila, 2011, Svanberg et al., 2013, Danielsson and Liss, 2012). Both modes have a lower variable transport cost per transported km and lower environmental impact, but also have higher fixed costs and require large freight volumes, terminals, and special infrastructure. The concepts of intermodal and co-modality highlight the opportunities in utilising the most ap- propriate traffic mode or combination of traffic modes to reach an efficient supply chain. In- termodal transport has been found to have less environmental impact than competing modes (Kreutzberger et al., 2003). Intermodality concerns the combination of different traffic modes in one transport chain; co-modality is the most appropriate use of each traffic mode (UN/ECE, 2001, European Commission, 2006). Both have received much political interest in recent years (European Commission, 2006). A further distinction can be made between intermodal transport and multimodal transport, where intermodal transport assumes the use of the same load unit through the chain while multimodal transport allows for the transhipment of loose cargo, e.g., bundles or chips. True intermodal transport for biofuels have rarely been studied in the scientific literature (Wolfsmayr and Rauch, 2014). All traffic modes have different characteristics and it is important to examine each traffic mode and terminal handling in the context of biofuel transport to understand the potential of each mode and their combinations. 154 The aim of this study is to investigate how the identified challenges affect the potential use of intermodal transport of biofuel for use in HPs. This study leads to a better understanding of the strengths and weaknesses of intermodal biofuel transport and have practical implications of anyone in the process of designing such systems. The study is set under Swedish conditions as the use of biofuel for HPs are particularly well established in Sweden (Björheden, 2006, Ericsson et al., 2004). Case A case study was performed concerning supplying a heating plant in Gothenburg, Sweden by rail. The plant is located in a residential area. At full operation, the plant consumes approxi- mately 17 GWh per week and is currently fuelled by wood residue chips, log chips, and stump chips. The plant is used for the base load in the district heating grid and is operated from early autumn to late spring. The plant has road access and access to the rail network via a non- electrified rail siding track. It is located next to an electrified major shunting yard and rail line. Local environmental regulations stipulate that chipping is not allowed at the terminal. Deliveries are allowed 24/7 but the plant tries to avoid weekend and night deliveries so as not to disturb the neighbours. The storage area is limited to 10 000 m3s, or roughly a 60 hour sup- ply (Friday evening to Monday morning). The plant is currently supplied by all-road, where all fuel is sourced locally. About 40 trucks deliver to the plant each day. The supplier is re- sponsible for the transport. All fuel is chipped road-side in the forest. Case methodology A potential rail system for the plant is designed and is subjected to a sensitivity analysis where key variables are changed to determine the key factors for successful intermodal transport. The potential rail system is designed by calculating the break even distance between road and rail transport. Within this region a search is made for potential terminals and sourc- ing locations, and a detailed rail system is designed. The rail system is then subjected to a number of scenarios, each one focusing on key characteristics of the system. Based on these scenarios, a “best feasible case” scenario is designed combining key characteristics from pre- vious cases. A modelling tool was developed in Microsoft Excel based on Flodén (2011). Much care was taken in finding good input data for the model, in particular the cost data. Six telephone inter- views and one e-mail interview with road (1), rail (1), and sea (e-mail) biofuel transport com- panies and a terminal company (1), forest company (1), and energy companies (2) were per- formed to further understand the operations. The interviews lasted about 60 min each and were recorded. Data was collected from the literature (Table ) and directly from the industry in Sweden. Real cost data were gratefully received from four industry actors in different parts of the supply chain. Some reported data for only parts of the operations. As can be expected, the cost estimates varied between the different sources. The cost data was refined by combin- ing the data from the industry and the literature, and checking the results against our own cal- culations of expected costs. This resulted in a data set for the cases containing a reasonable appreciation of the costs. Thus, the data used in the scenario does not represent the costs of any specific actor but can be viewed as an average cost level in the Swedish industry. All the selected data were independently validated with at least two industry representatives, while some data were validated with as many as five representatives. The selected cost levels were also validated with a reference group of biofuel industry actors from road, rail, power plant, and forest sectors. Costs were estimated in SEK, Swedish kronor, kr (2014: approx. 9 kr = 1€). See Appendix. 155 Table 5 Cost literature sources Data Literature sources Assumptions Biofuel densities Larsson and Nylinder (2014), COFORD (2003). Swedish conditions. Road costs Skogforsk (2011), Flodén (2011) Waiting times, etc. consid- ered. Empty returns. Rail costs Flodén (2011) Waiting times, etc. consid- ered. Emissions from Swe- dish electricity mix. Terminals Asmoarp (2013), Skogforsk (2011), Sommar (2010), Bäckström et al. (2009) Chipping Johansson and Mortazavi (2011), Eliasson et al. (2012), Lombardini et al. (2013) Bioenergiportalen (2013) Break-even distance As a first step in designing the base scenario the break-even distance between road and rail transport is of key importance to determine the minimum length of rail transport. A typical biofuel train is selected, consisting of 22 wagons (type Sgns), an electric engine (type Rd), and 45m3 load units with rotary unloading, transporting 2 300 MWh of logging residue chips. Rail has a 50km pre-haulage by road to the rail terminal, using a 93 MWh woodchip container truck. The train runs directly to the HP and is unloaded at the plant. Diesel shunting is used at both the terminal and the plant. The train is assumed to run three or five days per week, 26 weeks per year. The full round trip, including cost of the empty return transport, is included. Road transport is represented by a wood chip truck carrying 103 MWh, where 40% of the flow is transhipped at a road-road terminal. Twenty-three trucks are needed for the road transport and are assumed to return empty. Chipping is assumed to take place roadside in the forest in both systems. The calculations show the break even distance at 250 km for three days. Extending the train operations to five days per week pushes the break-even distance down to 180 km, showing the positive effect of high train utilisation. Figure 3 Costs for the biofuel system. From an environmental perspective, the advantage of the intermodal solution is clear. The intermodal solution produces significantly lower CO2 emissions compared to the all-road so- lution. The majority of this comes from the pre-haulage by road and chipping, as the rail transport has very low CO2 emissions. 156 Figure 4 Emissions for the biofuel system. Base scenario The base scenario design is based on interviews with the power plant and terminals in the area and has been validated with the plant. Among the major sourcing areas for biofuel in Sweden, the two closest areas above the break-even point are the regions of Småland and Dalarna. Due to the importance of a high utilisation of the train, a five day per week scenario is selected, operating three days a week to Småland (265km) and two days a week to Dalarna (471km). In Småland, logging residue wood chips are picked up, which is the most common biofuel in Sweden. In Dalarna bark is picked up. Dalarna is rich in wood industries, whose by-products are the second most common fuel (Awais and Flodén, 2014). In total, the system delivers 9.8 GWh weekly (58% of HP demand when operating at full capacity). Figure 5 The plant and sourcing locations. Smålan d Dalarna Gothenburg 157 Table 6 Base scenario characteristics Characteristics Train 20 wagons (type Sgns), an electric engine (type Rd), 60 45m3 load units with rotary unloading, 2 100 MWh of logging residue chips, or 1 750 MWh of bark. Terminal Electrified rail track in Småland with no diesel shunting. Diesel shunting in Dalarna. At both terminals the fuel is handled by wheel loader and the loading time is 4 hours. Road haul- age Road transport to the terminal is 40km by wood chip container truck carrying 93 MWh. Chipping takes place road-side in the forest for the wood residues. The bark requires no further chipping. Plant Diesel shunting to the plant. Unloading by a heavy forklift truck with a rotator. Unlading time is 4 hours. The cost per MWh for the analysed system is 99.95kr/MWh. This includes all activities, in- cluding chipping, road transport, terminal handling, etc. See Figure 6 for the cost allocation. Noteworthy are the high costs associated with chipping and the sending terminal. The cost for the rail part is 35.18kr/MWh (shunting, rail transport, and load units). Figure 6 Costs and emissions in the base scenario. From an environmental perspective, the CO2 emissions are 2.92 kg per MWh transported. The major sources are the road transport and chipping. Noteworthy is the low CO2 emission from the train transport, as a consequence of the clean Swedish electricity mix. Results A number of potential scenarios were evaluated. The following section divided the results according to Storage and terminals, Pre-treatment, Rail transport, Pre- and post-haulage, and Load units and transshipment. Storage and terminals Terminals are used for transshipment and storage, and this study has shown that the terminal costs can vary significantly due to operating practices and commodities handled at the termi- nal. Many companies also use existing round wood terminals for biofuel; in many cases these are old and already written off. Biofuel is perceived as a low-marginal product at the terminal 158 and companies state that biofuel is often only charged a marginal handling cost. Assuming that the road-rail terminal cost could be lowered to 10 kr/MWh, or roughly representing the variable costs of two handlings with a wheel loader, this would result in a 9% reduction of system costs. On a system level, this is particularly interesting in a comparison with all-road transport, as a significant share of all-road transport volumes already pass through a terminal (Björklund and Eriksson, 2013). Values range from 20% to 60% for individual companies (Enström et al., 2013). For an operator considering switching from all-road transport to inter- modal transport it should be noted that not all terminal costs in the base scenario are added costs: some are probably already incurred in an all-road system. Shunting costs are hard to estimate, as they are very situation-dependent for each terminal. Bäckström, et al. (2009) surveyed the shunting at a number of Swedish conventional inter- modal terminals and found that the time for shunting ranged from 20 minutes to 1 hour, not including administrative tasks and break tests. Assuming that the shunting at both the plant and terminals could be cut to a minimum amount of time with all equipment available at the terminal and minimum administration, this could result in a system cost reduction of 1% or 3% for the rail part (shunting, rail transport, and load units). A way of reducing the cost for shunting is to use the electric long haul engine. This removes the need for a separate diesel shunting engine but requires a special track layout and an electrified line. Electrified shunting reduces the costs by 0.4% for the system or 1% for rail. In comparison, assuming diesel shunt- ing at all locations in the base scenarios increases the costs by 0.3% for the system and 1% for rail. Pre-treatment The chipping position in the supply chain potentially has a large impact on transport costs. Chipping costs can be lowered by chipping at a terminal, but this will come at a cost for more expensive road transport due to the low density of logging residues. Changing the chipping location for the wood residues to the terminal results in a 1% reduction in the base scenario. However, this is very dependent on the distance to the terminal. If the distance is doubled to 80 km for the logging residue part, the costs are increased by 17%. From an environmental perspective, chipping at the terminal releases less CO2, but this is outweighed by less efficient road transport, resulting in a total increase in CO2 emissions of 21% for the first scenario and as much as 77% for the 80km haulage. Figure 7 CO2 emission and costs distribution when chipping at the terminal. The importance of the chipping location is made clearer by examining a system with only logging residue chips. Assuming that the bark in the base scenario is replaced by logging resi- 159 due, a comparison has been made between road-side chipping and terminal chipping, depend- ing on road distance to the terminal. A break-even point can be found at 48km road haulage, after which road-side chipping gives a lower cost. (See Figure 8.) The chipping equipment at the terminal gains its cost advantage by being larger, but therefore also requires larger vol- umes. Note that since about 80% of a pile of logging residues is air and the space at the termi- nal is limited, it is important to chip the biomass when it arrives to avoid filling the terminal. In comparison, 1 MWh of wood residues takes up 3.5 m3, while 1MWh of logging residue chips only take up 1.3 m3. Wood chips can also be piled higher, thus making more efficient use of each m2. Figure 8 Break-even point between road-side chipping and terminal chipping. The calculations assume that the fuel in both cases must pass a terminal, thus incurring the terminal costs. The break-even distance would be shorter in an all-road system, where the option to chip roadside also means that the terminal cost could possibly be avoided. A special case is round wood, as this almost always is sent to a terminal for chipping since most HPs are not allowed to chip at the plant. This makes these flows attractive from an intermodal point, since they already incur the terminal cost. Round wood also has a high density and thereby lower transport costs. Assuming that both the bark and logging residues in the base scenario are replaced by round wood that is transported by a timber truck and chipped at a terminal brings a 14% cost reduction. If, assuming that the biofuel would only be charged the marginal cost for the terminal operations of 10 kr/MWh, the cost would drop by 23%. Note that the price of the biomass is not included, which might differ between logging residues and round wood. Rail transport The choice of rail equipment has a significant effect on cost. Chaining to older equipment (RC4 engine and Lgns wagons) results in a system cost reduction of 4% and a rail cost reduc- tion of 12%. Due to weight restrictions on the wagons, this system carries slightly less biofuel (9.6GWh per week). Operating with old equipment also increases the risk of technical failures and disruptions. A safer system can be achieved by operating more modern engines. A new modern engine - e.g., TRAXX, which is equipped with ERTMS - brings a cost increase of 4% and 10% for system and rail, respectively. The TRAXX is also a stronger engine and can pull heavier trains. Utilising maximum load capacity (15.2 GWh per week) results in a system cost reduction of 17%. The use of the more expensive engine can balanced out against the in- creased loading capacity; however, such a large and heavy train cannot operate on all tracks and terminals, making this dependent on whether or not the extra capacity is needed and can be utilised. 160 Of particular importance is the utilisation of the rail engine. A rail engine has high fixed costs and higher utilisation brings a lower cost per hour. Assuming that the rail engine could find full employment outside the base scenario (i.e., weekends and summer), this would result in a reduction of system costs by 3% and 9% for rail. If both engine and rail wagons could be used similarly outside the system, the cost would reduce by 7%. For the rail part, this would be a 19% reduction. If it is further assumed that the load units could also find another use, the costs go down by 9% and 26%, respectively. Naturally, this is an extreme example, as a 100% utili- zation of the resources outside the system is not practically possible; however, it shows the importance of a high degree of utilization. Current operators state that they can find no use for wagons and load units during the summer months, but that the engine often has some alterna- tive use. Biofuel trains normally run empty on the return trip. Assuming full return flows in the rail system reduces the cost by 17%. The cost for the rail system is reduced by 50%. This assumes that the train is completely filled on the return, using the same load units and terminals, and that no detours are required. This is, of course, not realistic, but shows the maximum effect of gaining return flows. Return flows might also exist on the road side. If the road haulage trucks receive full return flows, then the costs are reduced by 7%. Full return flows on both trucks and train reduces the costs by 25%. Naturally, the possibility of return flows also exists in an alternative all-road transport system and would have a similarly large impact. As in all rail systems, larger trains bring lower costs per transported unit. The effects of train size on the current base scenario can be seen in Figure 9. Current infrastructure and regula- tions do not allow larger trains, but this has been disregarded in the calculation. However, extra engines have been added when necessary to be able to pull heavy trains. The peak in the curve represents one extra engine. Figure 9 The effect of train length. Similarly, changing the overall train utilisation has a large impact. The base scenario assumes operations five days per week. Assuming that the train could only operate three days per week to Småland results in a 17% cost increase. Similarly, assuming that the train would operate 7 days per week reduces the cost. An extended case is constructed assuming that the base case is complemented with two shorter runs on Saturday and Sunday to Varberg, 78 km south of Gothenburg, where wood chips are imported by ship. Costs for the ship and chipping are not considered. The extended scenario delivers 14.0 GWh per week at a 23% cost decrease. The effect of extending the operation to 7 days is particularly large since in the base scenario cal- culations are assumed that trains stand idle during the weekend Train utilisation is also affected by the length of the season during which the train is operated. Assuming that the train could be operated all year would reduce the costs by 8%. This would 161 not be realistic due to the limited need for heat during the summer, but even an extension of the season by one month would reduce the costs by 2%. (See Figure 10.) The extension could also be made possible by taking other complementing goods to nearby locations or increasing the summer storage at the plant. Note that an important factor in these calculations is how much outside use the train set has, particularly when operating only a few weeks in this sys- tem Figure 10 Length of season and costs. Rail transport is an inflexible mode of transport, where timetables are set long in advance and possibilities of deviations are limited. However, the fuel demand can vary depending on the outside temperature. The plant can operate at as low as 30% capacity before having to shut down for technical reasons. The demand reduction can be handled by either running the trains as normal, but loading less fuel, or by cancelling some trains. The characteristics of rail transport - with planning and contracts made long in advance are combined with the unpre- dictable weather and demand for heat - make most current rail set ups operate according to the first principle (or, if possible, redirecting to a different destination/customer). (See Figure 11.) System costs increase by 36% when the train is 50% full. Looking only at the costs for the rail part (shunting, train, and load units, but not loading/unloading), the costs increase by 87% per MWh. Figure 11 Increasing costs based on number of weeks with 50% full trains. Different types of fuel have different densities and energy content per tonne, thus resulting in different transport costs. Calculations have been made based on the base scenario, but with the same fuel on all trains and the maximum train capacity. The rail costs include rail 162 transport, shunting, and load units. The most expensive fuel to transport is sawdust with a low energy content. The different types of wood chips all have similar costs. Figure 12 Rail costs for different fuels. Pre- and post-haulage The road haulage to the terminal constitutes a large cost in the system. A haulage distance of only 10km cuts the cost by 11%, while a distance of 100km increases the cost by 22%. This shows the importance of keeping the haulage distance to the terminal short. In the base sce- nario, the train runs directly into the plant. However, most plants lack direct rail access and require a separate terminal with haulage by road for the last part to the plant, resulting in a 23% cost increase. (See Figure 13.) This assumes a terminal with diesel shunting located 20 km from the plant and road haulage by wood chip truck carrying 103 MWh wood chip residues or 72 MWh bark. The train is unloaded by a forklift truck with a rotator and loaded to the truck by a wheel loader. Figure 13 Distribution of costs and CO2 emissions. As with the haulage cost to the terminal, the distance from the terminal to the plant is also important. A 10km haulage increases the cost by 19% and a 50km haulage by 50%. The im- portance of keeping the haulage short to and from the terminals is further highlighted by look- ing at the combination of long and short haulage distances in both ends. A 10km haulage in both ends gives an 8% cost increase, while a 100km haulage gives an increase of 72%. This very high cost increase clearly shows the need to keep the distance short. 163 Figure 14 Effect of changed road distance to and from terminal. The lines represents the change in total costs and the bars the road costs share of total costs. Load units and transhipment There are many different types of load units that can be used for transport, as discussed in Falkenberg and Sökjer-Petersen (2014). One of the most common types of load unit used for biofuel transport is the traditional 40m3 wood chip container. The 40m3 container has less efficient unloading than the rotary container, thus extending the unloading time. Unloading from the train is normally made by a smaller forklift truck and the container is emptied by transferring it onto a switch body truck that tips the container. The extra handling of transfer- ring the container to the switch body truck increases the unloading time. In the rotary system, the truck turns the container upside-down with a rotator attached to the fork-lift truck and empties it directly. Average unloading of a rotary container is around 4 minutes, compared to around 7-8 minutes for the container system (Enström and Winberg, 2009). This impacts the turnaround time of the train as it extends the unloading time of a full train to 7-8 hours as compared to 4 hours and thereby limits the utilisation of the train and terminal. This could particularly be troublesome for a very large plant that requires several trains per day. Loading capacity on the train is reduced to 8.7 GWh per week, resulting in a 4% cost increase. How- ever, the smaller loading capacity reduces the train weight and also makes it possible to in- crease the number of wagons on the train to 22 without exceeding the maximum train weight. This makes it possible to deliver 9.6 GWh, which is close to the base scenarios of 9.8 GWh, at a 2% cost increase. The impact on the rail costs for the longer unloading time depends largely on cycle time for the train. In most cases, the train runs on a 24 hour cycle with one delivery per day, con- strained by the opening hours of the plant and terminal. Unless the transport distance is very short, it is not possible to make two cycles per day. As long as the extra 4 hour unloading time does not impact the 24h cycle, or prevents the train from making two runs per 24h cycle, the train costs will remain largely unchanged (exact costs will depend on the timetable). How- ever, as in the current case, if the longer unloading time makes the run to Dalarna impossible the cost will increase significantly. Assuming that the run to Dalarna is impossible and the train stands idle, the costs increase by 18%for the 40m3 system with extra wagons. The cost for the fork lift truck is significantly less for the smaller truck used in the 40m3 sys- tem. The fixed truck costs are about 3-4 times higher for the rotary system, making a high utilisation of the trucks important as the fork lift trucks are stationed at the plant. However, the smaller truck in the 40m3 system needs a switch body truck for unloading, but these are available on the open market and can be contracted per hour. The break-even point between 164 the unloading costs of two systems are at around 4 trains per week if operating half the year with trains of similar loading capacity and carrying only logging residue chips. Note that this only refers to the unloading process and not the total system. Figure 15 Unloading costs for rotary system and 40m3 using switch body trucks. A system comparison was made assuming a 300km transport with logging residue chips, with the same characteristics as the break-even calculation above: 40km pre-haulage and train de- livery at the plant. This shows that the rotary system gains competitiveness as the utilization increases. Figure 16 Comparison between a 300km transport with rotary system or 40m3 containers. A true intermodal system includes the transshipment of a detachable load unit between road and rail, thus transporting the fuel in the same load unit all the way to the plant. The ad- vantages of transferring the load units is more efficient transshipment at the terminal and that conventional intermodal terminals can be used. The drawback is that the loading capacity on the train is reduced as it is limited by the maximum allowed weight on the road. Also, more load units are needed since they are also used for the transport from the forest and for storage at the terminal while waiting for the train. If there is road haulage in one part of the chain then there is a need for at least double the number of load units on the train so there are load units ready to be loaded on the train, or the trains dwell time must be significantly increased. Con- sequently, the need is tripled if there is road transport at both ends. However, to maximize the transport from the forest there might be a need for even more containers. Each container truck normally uses six containers. When three full containers are picked up in the forest, three empty ones are dropped-off. By the time the truck has delivered the fuel and returns with 165 empty containers, the containers in the forest have been filled and are ready for pickup. This minimizes the waiting time for the truck and maximizes the use of chipping equipment, etc. in the forest, but requires extra containers. This further depends on if it is possible to unload the empty container from the truck in the forest. Switch body trucks can perform the unloading themselves, while other trucks and load units, including rotary containers, require external equipment such as fork lift trucks. These are not available in the forest; therefore, the system loses some of its effectiveness as it turns into a “hot” system that requires the simultaneous presence of both the truck and the chipper (or other loading equipment if the fuel is already chipped). If intermodal transshipment is used in the base scenario at the terminal, the cost remains al- most the same, with a 1% reduction. Looking at the different costs, the road haulage costs increase by 22% and the load unit costs by 100%. (See Figure 17.) However, the terminal costs decrease by 42%, which balances out the increased costs. Note that the terminal costs assume a conventional intermodal road-rail terminal that has high efficiency in transshipping containers. If a forest terminal is used for the container transshipment, the costs will likely by higher since the equipment utilization and experience in container handling will be lower. Terminal design is also not adapted to container handling. Road transport with the rotary con- tainers is further constrained in that they exceed the maximum width allowed on the road and therefore require a special permit (Falkenberg and Sökjer-Petersen, 2014). Extra costs and administration for this have not been included, nor have any extra costs in the forest due to the “hot” system. The true cost of the intermodal system is therefore probably higher. The high road transport costs also make the system more sensitive to longer haulage distances. (See Figure 18.) The higher road cost is caused by the decreased loading capacity per truck, as only two rotary containers can be carried per truck without violating maximum weight restrictions on the road. Figure 17 Distribution of costs and CO2 emissions with intermodal transshipment for the base scenario. 166 Figure 18 Effect of changed road distance to terminal with intermodal transshipment for the base scenario. Similar results are found for the 40m3 system with an increased number of wagons, at a 1% cost increase. In this case, the road transport costs are mainly unchanged, while the terminal costs are further increased due to the increased number of load units. Figure 19 Distribution of costs and CO2 emissions with intermodal transshipment for 40m 3 system. An interesting case is to only use intermodal transshipment in the receiving end of the chain. The transport from the forest into the sending terminal is already well established and effi- cient. In cities there is often no terminal suitable for bulk transshipment of biofuel; however, there is often conventional intermodal terminals suitable for container transshipment. Being able to use conventional terminals will therefore increase the number of possible terminals and thereby potentially reduce the road haulage distance between the terminal and the plant. In the current case there is a conventional intermodal terminal 6km from the plant. Further, the needs for load units are reduced since there is no need for extra load units for the forest operations. This system gives a cost increase of 12%, or 16% for the 40m3 system. Terminal handling costs are higher for the 40m3 system, since more load units are required and the rail transport costs are higher. The road transport costs and load units costs are lower, but this does not fully balance the higher train and terminal costs. 167 Summary of case results The tested cases are summarised in Table 7. The table is sorted according to cost, and an ap- proximate break-even distance with road is given based on a wood chip truck with logging residue chips, roadside chipping, and where 40% of the flow is transhipped at a road-road terminal. This is the same assumption as when the break-even between road and rail was pre- viously calculated. The results have also been plotted against their CO2 emissions in Figure 20. Cases with extreme values have been removed to give a more detailed overview. As can be seen, the cases with a high utilisation of resources give the lowest costs. Cases with long road haulage distances and more use of terminals give the highest cost. 168 Table 7 Summary of results 169 Figure 20 Change in costs and emission from the base scenario for tested cases without outliers. 1 Best feasible case scenario Based on the calculations above, a best feasible case scenario for the heating plant was de- signed. A low cost system should first have a high utilization of resources. A system operating 7 days per week would deliver 14.0 GWH, or 82%, of the plant’s need at full power. Considering that the plant does not always operate at full power and the risks associated with depending too much on one train, it is considered appropriate to choose a 5 day per week operation, as in the base scenario. Chipping round wood at the terminal has shown to be very efficient and is selected instead of logging residue chips for the part to Småland. A more modern rail engine would increase the loading capacity on the train and thereby lower the costs. However, con- sidering the current infrastructure, such large trains are considered unsuitable. Also, the deliv- ered volumes would increase to 15.2 GWh per week, or 89% of maximum need. An older engine with old wagons would also decrease costs, but at a higher risk of technical failures and disruptions. The older engines are therefore not selected due to the need for a reliable system. The rotary containers are considered the most effective and are selected. Return flows would also reduce the costs, but are considered very hard to get and are therefore not includ- ed. Road haulage distances should be kept as short as possible. However, for this calculation it is assumed that the distance in the base scenario remains the same, as this is the average distance today. S several actors in the wood processing industry have rail access at their plants. Selecting to source from an actor with rail access could lower the transport cost. It is assumed that an actor with rail access is selected. Increased utilization of the train set outside the system would further decrease the costs, but it is hard for the power plant to influence this decision. However, extending the season for the train could be possible, but would largely depend on the outside temperature, which is impossible to predict. In this scenario, a slight increase of the season by two weeks to 28 weeks per year is assumed. Other aspects of the scenario remain as in the base scenario. Figure 21 Distribution of costs and CO2 emissions for the best feasible case scenario. The best feasible case scenario gives a cost decrease of 22% with CO2 emissions of 1.79kg/MWh. A rough break-even point against all-road transport is at 106km. Reducing CO2 emissions are of high importance, although this seldom carries an economic cost. The emission in the best feasible case solution mainly relates to the chipping and road haulage. The chipping emission can potentially be reduced by replacing the diesel powered chipping by electric powered chipping. Electric powered chippers exist but require large vol- umes to be economical. Often they only exist at large plants and sometimes at very large ter- 2 minals. A switch to electric chipping would not only reduce the emissions but also the costs, assuming a high utilization of the chipper. In the current scenario, the emissions would be drastically reduced if the terminal could be equipped with an electric chipper. In the best fea- sible case scenario, costs would be reduced by 32% and emissions to 1.18 CO2 kg/MWh. However, very few terminals would have the volumes required for an electric chipper. Road haulage should try and be as short as possible to keep CO2 emissions at a low level. The emis- sions could also be further reduced by alternative fuels. Diesel powered shunting should, as far as possible, be replaced by electric shunting and excess shunting should be avoided, for example by extending rail siding so that the trains do not have to be split. For a true intermodal transport solution where the same load units are transshipped through the chain, it is advisable to avoid bringing the load units all the way into the forest. This re- quires several extra load units or interferes with an already effective system. Also, if larger units such as the rotary containers are used, there is also a practical issue of limited space for handling and marshalling load units at the roadside and possible extra handling equipment needed. A better option is to use bulk loading at the sending terminal by bringing the contain- ers all the way to the plant at the receiving end. This comes at an increased cost compared to train deliveries directly to the plant (19%), but less than if bulk transshipment is used at the receiving terminal (23%). The cost advantage can be further increased if the conventional intermodal terminal is closer to the plant than the bulk terminal. This is also likely since there are more conventional intermodal terminals in and around the cities than bulk terminals. Note that this assumes that the rotary containers are transported by road, which requires a special permit. Also, the terminal costs are very important here. An efficient bulk terminal, in com- parison with a less efficient conventional intermodal terminal, easily gives the bulk option the lowest cost. Analysis Ultimately, the competitiveness of intermodal transport depends on where the fuel can be sourced and at what price. Either the plant is so large that fuel cannot be sourced locally or it must be sourced above the break-even distance for intermodal transport and/or type of fuel and price variations between regions, where a better price of fuel can be obtained from a far- away region. However, a properly designed intermodal transport system can shorten the break-even distance and make intermodal transport competitive on shorter distances, which is desirable due to the lower emissions from intermodal transport. A 10% cost change in the base scenario equals an approximately 33km shift in the break even distance. A successful intermodal transport system for biofuel must meet the five key logistics challenges identified (Awais and Flodén, 2014): seasonal variability, storage, chipping, low density, and dependen- cy on political policies. Seasonal variability Rail as a transport mode requires large volumes and a high utilisation of the equipment to be economically efficient. Also, the need for transhipment and road haulage to/from terminals makes it impossible for intermodal transport to be competitive for smaller plants that can source all their fuel locally with a short road transport. Only the larger plants may therefore be candidates for intermodal transport. The plant studied in this case at 17 GWh/week is on the smaller size for intermodal transport, due to seasonal variability. Intermodal is attractive when operating at full capacity but, e.g., a 50% full train increases the costs by 36%. The seasonal variability also poses a challenge for road transport, but to a less extent due to the more flexi- ble road system where capacity can be more easily adapted and outside transport assignments 3 more easily found. To overcome the seasonal variability intermodal transport must be applied at large plants where the train can be run at full capacity during most of the season. The ef- fects can also be reduced by intensifying attempts to find alternative uses for the equipment outside the season. Storage Storage is often performed at the plant or at the terminals. The use of storage at terminals is an advantage for intermodal transport. In a comparison with an all-road system it is important to remember that several of the activities are performed both in an all-road system and in an in- termodal system. For example, a large share of the all-road volumes already pass a terminal, and in some cases a terminal with rail access. Values range from 20% to 60% of the flows for an individual company to pass a terminal (Enström et al., 2013). These flows therefore al- ready incur a terminal cost, which makes the additional terminal cost low for loading the fuel on a train instead of on a truck at the terminal. Intermodal transport is thus more competitive against flows that already pass a terminal. This is particularly true for round wood, which al- most always passes a terminal. Large costs are associated with terminal activities. It is important to note that the costs associ- ated with in particular terminals can vary significantly. A very wide range of cost estimates have been found in the literature and in interviews with industry representatives. Thus, select- ing the right terminal is a key issue for a successful intermodal system. A related issue is also how the costs at a terminal are shared between different types of wood using the same termi- nal, e.g., pulp wood and biofuel. Chipping Chipping is most often performed road-side in the forest. However, whenever possible, inter- modal transport should try and utilise potential cost savings from using a terminal, such as the lower chipping costs at a terminal compared to road side chipping. For shorter pre-haulage distances, logging residues should be brought in to the terminal and chipped there, preferably with an electric chipper. Low density The low density of biofuels points to the importance of keeping the transport costs low. It can be concluded that a high utilisation of the resources is the most important aspect in keeping the costs down. Also, road distances to and from the terminals should be kept as low as possi- ble. In particular, the very long pre/post-haulage distances of 80-100km gives very high costs for the system. Options with rail access directly at the plant or directly to a forest industry are preferred. For example, adding a terminal in the receiving end and 100km road haulage to the plant increases the cost twice as much as the cost savings gained from finding full return loads for both trains and trucks in the base scenario. Thus, avoiding terminals and keeping road dis- tances short is the most important aspect. Fuels that can be picked up directly at a terminal without road haulage have the greatest potential, for example forest industry by-products from industries with rail access. Political policies Policies impact the intermodal biofuel system on three levels: local, industry, and national. On a local level, the plants are often municipality owned and subject to political regulations. In- termodal transport would be favored by political instructions for the plants to place a higher importance on CO2-emissions. Currently, most plants have no specific requirements on transport emissions. Also, local permits, such as whether or not night-time deliveries are al- 4 lowed, impact the system and its flexibility. Since intermodal transport in its nature is less flexible, flexible local regulation improves the system flexibility and thereby its competive- ness. On an industry level, political policies impact the competitiveness of the industry as whole; e.g., by taxing non-renewable energy, electricity certificates, etc. Reduced competitiveness for the industry brings down the fuel volumes needed, making it harder for intermodal transport. On a national level, political decisions also impact the general competitiveness of transport modes; e.g., the level of road taxes, fuel taxes, infrastructure fees, etc. Any political incentive in favor of intermodal transport in general will naturally also impact the potential for inter- modal biofuel transport. Conclusions The possibilities for intermodal transport to overcome biofuel logistics challenges are summa- rised in Table 8. Table 8 Logistical biofuel challenges and the intermodal options Challenge Intermodal options Seasonal variability Focus on large plants to fill the train all season. Storage Utilise the intermodal terminals and focus on flows that al- ready pass a terminal. Select an efficient terminal. Chipping Utilise the lower chipping costs at the terminals. Low density High resource utilisation. Short pre-/post-haulage distance by road. Pick-up/deliveries directly to the plant. Dependency on policies Influence local and national politicians for flexible regulations and support for intermodal transport. Further, the competitiveness of intermodal transport can be improved by finding alternative sourcing locations where lower fuel costs can be obtained, rather than by sourcing locally. It is evident that intermodal transport is a system requiring large volumes and high resource utilisation. A potential for intermodal transport of biofuels exists for the larger plants sourcing from long distances, but intermodal transport is not competitive for smaller plants with shorter sourcing distances. 5 Appendix Input data Density kg/loose m3: Bark 350, Log chips 271, Logging residue chips 295, Saw residue chips 300, Sawdust 30, Stump chips 288, Whole three chips 300, Wood residue chips 225, Logg 367 Energy MWH/m3: Bark 0.65, Log chips 0.79, Logging residue chips 0.78, Saw residue chips 0.65, Sawdust 0.58 Stump chips 0.77, Whole three chips 0.8, Wood residue chips 0.8, Logg 1.07 Truck cost considering empty return, kr/km: Wood chip container truck 29.57 kr, Wood chip truck 30.77 kr, Forest residue truck 34.45 kr, Timber truck 29.57 kr Truck emissions considering empty return, CO2 kg/km: Wood chip container truck 2.95, Wood chip truck 3.02, Forest residue truck 3.13, Timber truck 2.24 Rail engine annual fixed costs: Rd 2 149 868 kr, Modern engine 3 426 462 kr Rail engine variable costs, kr/km: Rd 16.44 kr, Modern engine 19.08 kr Rail staff cost: 692,79 kr/train hour Wagon annual fixed costs: Lgns 41 221 kr, Sgns 63 119 kr Wagon variable costs, kr/km: Lgns 0.23 kr, Sgns 0.31 kr Load unit annual costs: Rotary container 13 100 kr, 40m3 container 11 521 kr Terminal costs, kr/MWh: Road to road 14 kr, Road to rail 19 kr, Rail to road 21 kr Fork lift truck annual fixed costs: Heavy truck with rotator 764 141 kr, Light truck 192 920 kr Fork lift truck staff cost: 310 kr/hour Chipping cost, logging residues, kr/MWh: Mobile chipper at terminal 23 kr, stationary electric chipper 7 kr, roadside chipping 40 kr Chipping cost, Logg, kr/MWh: Mobile chipper at terminal 20 kr, stationary electric chipper 5 kr, roadside chipping 35 kr References ASMOARP, V. 2013. Terminalstrategier för skogsflis på Södra Skogsenergi (Terminal strategies for wood chips at Södra Skogsenergi). Uppsala: SLU - Swedish University of Agricultural Sciences. AWAIS, F. & FLODÉN, J. 2014. Logistic requirements and characteristics of the Swedish wood biofuel industry. Sustainable Intermodal Supply Systems for Biofuel and Bulk Freight, WP2 Report, University of Gothenburg. BIOENERGIPORTALEN. 2013. Skogsbränslebranschen är ologisk, http://www.bioenergiportalen.se/?p=6630. Bioenergiportalen. BJÖRHEDEN, R. 2006. Drivers behind the development of forest energy in Sweden. Biomass and Bioenergy, 30, 289-295. BJÖRHEDEN, R., ELIASSON, L., IWARSSON, M., VON HOFSTEN, H., ENSTRÖM, J. & JOHANNESON, T. 2010. Efficient forest fuel supply system. In: THORSEN, Å., BJÖRHEDEN, R. & ELIASSON, L. (eds.). Uppsala: Skogforsk. 6 BJÖRKLUND, L. & ERIKSSON, U. 2013. Virkesmätning vid värmeverk - resultat från en inventering utförd 2012. Sundsvall: SDC Skogsnäringens IT-företag. BÄCKSTRÖM, S., BOHLIN, M., FRANZEN, U. & JONSSON, P. 2009. Miljökalkyler för intermodala transportkedjor - Detaljerad beräkningsmetodik och relevanta schablonvärden. SIR-C Swedish Intermodal Transport Research Centre. WSP Analys & Strategi. CARLSSON, D. & RÖNNQVIST, M. 2004. Supply chain management in forestry––case studies at Södra Cell AB. European Journal of Operational Research, 163, 589-616. COFORD 2003. Maximising the Potential of Wood use for Energy Generation in Ireland. Horsham: COFORD. DANIELSSON, B.-O. & LISS, J. E. 2012. Branschintervjuer, tankar om skogsbränslets framtid. . : Projekt SWX-Energi ELIASSON, L., GRANLUND, P., JOHANNESON, T., VON HOFSTEN, H. & LUNDSTRÖM, H. 2012. Flisstorlekens effekt på en stor skivhuggs bränsleförbrukning och prestation (Effects of target chip size on performance, fuel comsumption and chip quality for a large disc chipper). Uppsala: Skogforsk The Forestry Research Institute of Sweden. ENERGIMYNDIGHETEN 2012. Energiläget 2012. Eskilstuna: Energimyndigheten. ENSTRÖM, J., ATHANASSIADIS, D., ÖHMAN, M. & GRÖNLUND, Ö. 2013. Framgångsfaktorer för större skogsbränsleterminaler (Success factors for larger energy wood terminals). Uppsala: Skogforsk The Forestry Research Institute of Sweden. ENSTRÖM, J. & WINBERG, P. 2009. Systemtransporter av skogsbränsle på järnväg. Uppsala: Skogforsk The Forestry Research Institute of Sweden. ERICSSON, K., HUTTUNEN, S., NILSSON, L. J. & SVENNINGSSON, P. 2004. Bioenergy policy and market development in Finland and Sweden. Energy policy, 32, 1707-1721. ERIKSSON, L. N. 2008. Comparative analyses of forest fuels in a life cycle perspective with a focus on transport systems. Resources Conservation and Recycling, 52, 1190-1197. EUROPEAN COMMISSION 2006. Keep Europe moving - Sustainable mobility for our continent, Mid-term review of the European Commission’s 2001 Transport White Paper. Brussels: European Commission. FALKENBERG, A. & SÖKJER-PETERSEN, S. 2014. Transport Technology for Sustainable Intermodal Transports of Biofuel, WP4 report. Höganäs: Mariterm. FLODÉN, J. 2011. Rail freight costs - Some basic cost estimates for intermodal transport Göteborg: School of Business, Economics and Law at the University of Gothenburg, Swedish Intermodal Reserach Centre, Sir-C, http://hdl.handle.net/2077/25877. FROSCH, M. & THORÉN, P. 2010. Järnvägstransport av biobränslen. VÄRMEFORSK Service AB. GOLD, S. & SEURING, S. 2011. Supply chain and logistics issues of bio-energy production. Journal of Cleaner Production, 19, 32-42. HAMELINCK, C. N., SUURS, R. A. A. & FAAIJ, A. P. C. 2005. International bioenergy transport costs and energy balance. Biomass & Bioenergy, 29, 114-134. JOHANSSON, J. & MORTAZAVI, R. 2011. Transport av skogsenergisortiment - Företags- och samhällsekonomiska kostnader. SWX-Energi. KREUTZBERGER, E., MACHARIS, C., VEREECKEN, L. & WOXENIUS, J. Is intermodal freight transport more environmentally friendly than all-road freight transport? A review. The NECTAR conference no. 7, June 13-15 2003 Umeå, Sweden. LARSSON, F. & NYLINDER, M. 2014. WeCalc - Wood Energy Calculations [Online]. Uppsala: SLU - Swedish University of Agricultural Sciences. Available: http://woodenergy.sites.djangoeurope.com/conversion/ [Accessed March 2014]. 7 LINDHOLM, E.-L. 2010. Energy use and environmental impact of roundwood and forest fuel production in Sweden, Uppsala, Doctoral Thesis, Swedish University of Agricultural Sciences LOMBARDINI, C., GRANLUND, P. & ELIASSON, L. 2013. Bruks 806 STC – Prestation och bränsleförbrukning (Performance and fuel consumption of the Bruks 806 STC chipper). Uppsala: Skogforsk The Forestry Research Institute of Sweden. MAHMUDI, H. & FLYNN, P. C. 2006. Rail vs Truck transport for biomass. Applied biochemistry and biotechnology, 129-132. RENTIZELAS, A. A., TOLIS, A. J. & TATSIOPOULOS, I. P. 2009. Logistics issues of biomass: The storage problem and the multi-biomass supply chain. Renewable & Sustainable Energy Reviews, 13, 887-894. SKOGFORSK 2011. FLIS 4.0 Flexibel lathund för interaktiv systemanalys. Uppsala: Skogforsk The Forestry Research Institute of Sweden. SVANBERG, M. & HALLDÓRSSON, Á. 2013. Supply chain configuration for biomass-to- energy: the case of torrefaction. International Journal of Energy Sector Management, 7, 65-83. SVANBERG, M., OLOFSSON, I., FLODÉN, J. & NORDIN, A. 2013. Analysing biomass torrefaction supply chain costs. Bioresource Technology, 142, 287-296. SWEDISH FOREST AGENCY 2013. Swedish Statistical Yearbook of Forestry. Jönköping: Swedish Forest Agency, Skogsstyrelsen. SVENSK FJÄRRVÄRME. 2014. Svensk fjärrvärme (Distric Heating in Sweden) [Online]. Svensk Fjärrvärme, Swedish District Heating Association. Available: http://www.svenskfjarrvarme.se [Accessed February 2014]. TAHVANAINEN, T. & ANTTILA, P. 2011. Supply chain cost analysis of long-distance transportation of energy wood in Finland. Biomass and Bioenergy, 35, 3360-3375. TIMILSINA, G. R. & SHRESTHA, A. 2011. How much hope should we have for biofuels? Energy, 36, 2055-2069. UN/ECE 2001. Terminology on Combined Transport. New York and Geneva: United Nations (UN) & Economic Commission for Europe (ECE). WOLFSMAYR, U. J. & RAUCH, P. 2014. The primary forest fuel supply chain: A literature review. Biomass and Bioenergy, 60, 203-221.