(587ak) Linking Optimization and Simulation for Improved Biorefinery Supply Chain Design

Much research has been dedicated to chemical process modeling of various biofuels utilizing a variety of conversion techniques at a biorefinery. Additionally, supply chain modeling has been done to determine the optimal costs and configurations for delivery of feedstock to the biorefinery and products to consumers in the market. These models often times lack both the ability to assess the long-term viability of the supply chains and do not adequately represent the uncertainty and variability inherent to biomass based feedstock and energy markets. A much better perspective can be gained by dynamically modeling the long term performance of the optimal supply chain configuration determined through mixed integer linear programming.          

This presentation describes a discrete event simulation model created to combine inputs generated from process simulation and supply chain optimization models. Supply chain uncertainty is captured in the models with historically based probability distributions for biomass supply generation as well as for fuel product demand. Performance of the supply chain is evaluated through the net present economic value generated. According to the simulation, the case study supply chain for the Jackson Purchase Region of Kentucky (USA) does not show profitability over a 10 year operating period with the chosen chemical processes. However, key insights into the nature of the supply chain are revealed which provide valuable information relevant to potentially improving the profitability of similar supply chains. Benefits of simulation modeling in contrast to mixed integer linear programming alone are demonstrated and future applications of the model are discussed.