Designing a Resilient Biorefinery System under Uncertain Agricultural Land Allocation

Abstract

Abstract Agricultural residues are excellent feedstock for lignocellulosic biorefineries. However, the land allocated to various crops in a region can vary annually, thus impacting the feedstock availability for biorefineries. This work provides an optimization framework that considers uncertainty in land allocation for designing a biorefinery system that is resilient to such changes. A recently proposed decomposition-based approach is utilized to perform stochastic optimization, and the resulting design was compared to a deterministic design that considered mean land allocation. Lignocellulosic ethanol production for the state of Maharashtra, India, was taken as a case study, and the performances of both designs were evaluated on a set of 100 random land allocation instances. The resilient design had a smarter feedstock procurement strategy which resulted in a significant decrease in variation of feedstock procurement and transportation expenses. As a result, the variation in ethanol cost was 4% for the resilient design, as compared to 11% for the deterministic design.