(190r) Optimal Planning Under Uncertainty for a Supply Chain Focused on Residual Biomass Conversion Using Geographic Information Systems and Mathematical Programming

Biomass is a renewable resource with attractive characteristics for manufacturing many types of products such as biofuels and generation of energy. One of the main reasons to use residual bioresources for biofuels production is that biomass can capture a significant part of emissions during its growing. In this regard, various studies have addressed diverse important features such as development of technologies for biomass treatment and processing, schemes to determine location of biorefineries and harvesting sites, models to evaluate the economic and environmental impacts, approaches capable of considering the associated dynamic in biomass and products storage and supply, between others. However, the most of the works have omitted other important issues such as location of the potential supply chain’s nodes considering geographic constraints as well as the uncertainty associated with the residual biomass availability and price over time, mainly because this may increase drastically the number of options, variables and constraints to be taken into account. Therefore, this paper presents an approach for the optimal planning of a supply chain focused on biofuels production from agricultural residues, which involves a methodology to locate potential sites for biomass extraction considering environmental, social and geographic limitations using a tool based on geographic information systems, and a formal model based on mathematical programming. It is important to mention that planning problem for any supply chain consists on determining raw materials, products, processing technologies, processed and produced amounts, facilities’ location, among other issues to obtain diverse benefits. On the one hand, mathematical model is a Mixed Integer Linear Programming Problem and consists on mass balances for all considered biomass types, constraints for biomass availability and product demand, cost equations and functions to evaluate environmental and social impacts. On the other hand, the geographic constraints are the protected areas exclusion, terrain’s slope, wetland zones’ location, distance to water bodies, distance to human communities, frequency of weather phenomena like torrential rain and hurricanes, and distance to transportation and electrical infrastructure. Additionally, the proposed approach includes the variation in the biomass availability and price over time through a geoprocessing model by generating several scenarios based on historical information. The method was tested for a nationwide case study in Mexico to determine potential locations for biomass suppliers, processing plants, consumption regions and processing technologies to obtain biofuels from agricultural residues. The case study contemplates the location and production of several agricultural residues such corn stover, sugar cane bagasse, sorghum straw, residue from agave and wheat straw to produce diverse products such as Fisher Tropsch liquids, ammonia, formaldehyde, methanol, among others via different gasification alternatives. Results show the variation of potential locations for biomass suppliers to be used in the proposed biomass processing system, while the availability of bioresources changes. Furthermore, results illustrate the influence of the type of raw material in the location of these suppliers. The geoprocessing model was useful to identify potential locations to be taken into account in the subsequent supply chain planning problem decreasing considerably the number of options to choose accounting social, environmental and geographic issues, for instance for the case of corn stover the considered initial municipalities were 2337, nevertheless the geoprocessing model can reduce them to around 200 municipalities with high potential to install process facilities and biomass suppliers to be included in the mathematical model for optimal planning.