Coupled Optimization and Simulation for Multi-Biomass Source-to-Biorefinery Supply Chain Modeling and Analysis

As cost associated with fossil fuels rise due to various environmental, social, and economic factors, it has become prudent to examine the possibility of exploiting alternative sources of raw materials for the production of energy and chemicals. Biomass, of various forms, has been proposed as an option to fill this need for alternative raw materials. Several problems with biomass as an alternative to fossil fuel derived materials instantly arise: low bulk energy density, the resulting high transport costs, seasonal variability of biomass supply, and the risks inherent to relying on a largely agriculturally based system are several examples. To begin to realize energy and chemical production with biomass as the feedstock on an economic scale, supply chain optimization and a robust means of simulating variation over long periods of time is critical.

Due to a relatively low energy density of the given biomass feedstocks, the majority of the biorefinery supply chain costs occur in the process of raw material transportation. By optimizing this aspect of the biorefinery supply chain, given the plant and potential feedstock source locations, the cost associated with raw material transportation can be minimized. In order to examine the potential longevity and general robustness of the output from the optimization model, simulation is a valuable tool. In this way, the effects of perturbations on the various feed stock supplies due to weather variability, crop yield variability, or other factors can be examined and analyzed.

This presentation focuses on the development of a coupled system comprised of an optimization model for the upstream portion of biorefinery supply chains in Western Kentucky and a discrete-event simulation program. The optimization model minimizes the transportation costs associated with moving biomass feedstock from centralized storage facilities to the biorefinery location. Multiple feedstocks available in the region under consideration including chicken litter, forest residue, and corn stover were taken into consideration. Various alternative shipping methods as well as a centralized versus a distributed production facility layout are considered as well. The discrete-event simulation model allows for the examination of the robustness of the determined optimal supply chain design. This combination of optimization and simulation for the examination of multi-biomass source-to-biorefinery supply chains could potentially serve as a template for future work in biorefinery supply chain model development and decision-making.