Uncovering optimal biorefinery designs for agricultural waste: Insights from superstructure optimization under uncertainty

Abstract

This study explores the application of superstructure optimization to guide early-stage biorefinery design decisions under uncertainty. A case study on processing agricultural waste (potato peel waste) demonstrates the method's capacity to identify economically viable and robust pathways from two perspectives: waste owners and stakeholders invested in emerging technologies. For waste owners, stochastic optimization revealed that converting potato peel waste into animal feed is the most robust and economically attractive pathway, achieving an expected profit of 61.2 €/t of waste with minimal risk exposure. For technology developers, global sensitivity analysis of a polyhydroxyalkanoate (PHA) production route highlighted critical parameters for improvement, including PHA extraction yield and separation efficiency, to enhance economic performance and outcompete alternative production routes. The work includes enhancements to the OUTDOOR software, integrating uncertainty optimization and sensitivity analysis, to provide a comprehensive framework for evaluating design alternatives. This structured approach facilitates the identification of robust biorefinery configurations in the early design stage. It also allows stakeholders in emerging technologies to benchmark their processes against alternative designs, thereby identifying critical parameters for optimization