The purpose of this dissertation is to provide an effective approach to design the supply chain (SC) of the cellulosic biofuel industry in order that it will support and accelerate the successful commercialization of the cellulosic biofuel industry. The methods of approach to this problem are (1) to assess the state-of-the-art biofuel SC studies, (2) to provide a decision support tool based on a mixed integer programming (MIP) model for the cellulosic biofuel supply chain design problem (BSCP), (3) to devise an exact solution method to solve large-scale instances of BSCP, (4) to evaluate a biomass logistics system based on biomass modules, by using new simulation elements for new machines, and (5) to compare several biomass logistics systems based on biomass module, bale, and silage, using simulation models. The first part of this dissertation broadly reviews the literature on biofuel SCs, analyzing the state-of-the-art biofuel and petroleum-based fuel SC studies as well as relating generic SC models that have been published over the last decade to the biofuel SC (An et al., 2010a). The resulting analysis proposes fertile opportunity for future research to contribute to improving biofuel SC. The second part of this dissertation formulates BSCP as a MIP model, which is a time-staged, multi-commodity flow, network design problem with an objective of maximizing profit (An et al., 2010b). The model prescribes strategic level decisions (i.e., facility locations, capacities, and technology types) as well as plans for transportation routes and material flows (i.e., quantities produced, stored, and transported) in each time period. A case study demonstrates managerial use in application to a region in Central Texas. The third part of this dissertation provides an exact solution method to solve BSCP. An embedded structure can be transformed to a generalized minimum cost flow problem, which is used as a sub-problem in a CG approach. This study proposes a dynamic programming algorithm to solve the sub-problem in O(m), generating improving path-flows. To accelerate branch-and-bound (B&B) search, it develops an inequality, called the partial objective constraint (POC), which is based on the portion of the objective function associated with binary variables. The fourth part of this dissertation evaluates a biomass module system, which is a conceptual logistics system based on large packages of chopped biomass with sufficient size and density to provide maximized legal highway loads and quick load/unload times. The last part of this dissertation evaluates economic benefits of the biomass module system, comparing it to bale and silage systems.

The global energy requirement is increasing at a rapid pace and fossil fuels have been one of the major players in meeting this growing energy demand. However, the resources for fossil fuels are finite. Therefore, it is essential to develop renewable energy sources like biofuels to help address growing energy needs. A key aspect in the production of biofuel is the biomass logistics chain that constitutes a complex collection of activities, which must be judiciously executed for a cost-effective operation. In this thesis, we introduce a two-phase optimization-simulation approach to determine tactical biomass logistics-related decisions cost effectively in view of the uncertainties encountered in real-life. These decisions include number of trucks to haul biomass from storage locations to a bio-refinery, the number of unloading equipment sets required at storage locations, and the number of satellite storage locations required to serve as collection points for the biomass secured from the fields. Later, an operational-level decision support tool is introduced to aid the "feedstock manager" at the bio-refinery by recommending which satellite storage facilities to unload, how much biomass to ship, how to allocate existing resources (trucks and unloading equipment sets) during each time period, and how to route unloading equipment sets between storage facilities. Another problem studied is the "Bale Collection Problem" associated with the farmgate operation. It is essentially a capacitated vehicle routing problem with unit demand (CVRP-UD), and its solution defines a cost-effective sequence for collecting bales from the field after harvest. / Master of Science

This dissertation focuses on supply chain design and logistics problems with emphasis on biomass logistics and routing problems. In biomass logistics, we have studied problems arising in a switchgrass-based bio-ethanol supply chain encountered in the Southeast, and a corn stover harvest scheduling problem faced in the Midwest Unites States, both pertaining to the production of cellulosic ethanol. The main contributions of our work have been in introducing new problems to the literature that lie at the interface of the lot-sizing and routing problems, and in developing effective exact algorithms for their solution. In the routing area, we have addressed extensions of the well-known traveling salesman and vehicle routing problems. We have proposed new formulations and have developed exact algorithms for the single and multiple asymmetric traveling salesmen problems (ATSP and mATP), the high-multiplicity asymmetric traveling salesman problem (HMATSP) and its extensions, and the fixed-destination multi-depot traveling salesman problem with load balancing (FD-MTSPB). Furthermore, we have introduced a new strategy to reduce routing cost in the classical vehicle routing problem (VRP). / Ph. D.

Increasing differences in United States energy consumption and production has influenced the passing of legislation for biomass fuel production. To determine feasibility of energy crops for alternative fuels, research is needed to investigate dry matter yield over an extended harvest season; physical characteristics need to be described for potential harvesting problems; chemical characteristics described to identify selective harvest potential, optimal harvest timing, losses during harvest and storage; various harvest techniques investigated to identify potential cost savings; and impact of various storage techniques on quantity and quality of deliverable biomass. This study investigated the use of two sorghum varieties as a potential bioenergy feedstock where 20 ha were planted for three years. Standing crop samples were collected from August through January to document changes in dry matter yield, moisture, height, fiber content, proximate and ultimate analysis. The sorghum was cut and conditioned – as a two-cutting ratoon or single-cutting – using various mower-conditioners and windrow samples taken daily to determine best method of field drying, quantify dry matter loss and soil entrainment. Two storage methods were utilized – baling with wrapping in a tubeline, and chopping and compressing in bag using a modified cotton module builder – to determine best method of storage for reduced dry matter loss. The optimal time of harvest for maximum dry matter occurred with the November once-cut where 30 Mg ha^-1 was documented, but comparable yields were observed with the two-cutting scenario. Fiber content increased with maturity, peaked, and declined, while ash content and moisture decreased with maturity. The achievement of 55% moisture in January shows field curing to be necessary for transportation at any significant distance, but soil entrainment – as measured by ash concentration – was not found to be a significant problem after conditioning, multiple windrow inversions, and harvesting. The geometric mean length of particle was determined to be 1.4 to 3.7 times lower than nominal chop length, indicating potential cost savings in comminution. Dry matter loss estimates during storage proved difficult due to mobility of moisture throughout the packages, where losses were documented up to 40%. Module packages tended to have lower dry matter and constituent losses than bales.