(64f) Optimal Design of Biomass Processing Networks for Biofuel Production Under Uncertainty

Bio-fuel represents a promising candidate for renewable energy. It presents the biomass industry with exciting opportunities as well as significant challenges. One of the challenges for the emerging industry concerns a very high level of uncertainty around feedstock costs and supply amounts. The uncertainty makes the assessment of the economics and investment decisions very difficult. This study focuses on optimal design of biomass supply chain networks for bio-fuel production under significant uncertainties in biomass feedstock availabilities and their acquisition costs. Biomass yields are extremely sensitive to factors like the weather and supply costs are also affected by emergence of new industries competing for the same types of resources. Designing a processing network robust to uncertainties in price and availability are critical to assess and reduce the risk associated with investment in the bio-fuel industry. It is desirable to account for all possible future scenarios in designing and analyzing such a network. The supply chain network we study covers the Southeastern region of the United States and includes biomass supply locations and amounts, candidate sites and capacities for two kinds of fuel conversion processing, and the logistics of transportation from the locations of forestry resources to the conversion sites and then to the final markets. A two stage stochastic program is formulated and solved given the data for possible scenarios and their probabilities. The first stage decisions (called ?here-and-now decisions?) are the capital investment decisions including the locations and capacities of the processing plants, and the second stage decisions (called ?recourse decisions?) are the material flows from biomass sites to processing sites and eventually to the final markets. The MILP solves for the optimal numbers, locations, and sizes of the conversion processing plants, as well as the amounts of biomass, intermediate products, and final products for each scenario in order to satisfy to final market demands, with the objective of maximizing the expectation of the overall profit. The robustness of the nominal design (for a single nominal scenario) vs. the robust design (for multiple scenarios) is analyzed under various scenarios considered.