A multistage stochastic mixed integer optimization model for the design of a renewable fuel supply chain network under demand uncertainty

Computers & Chemical Engineering, vol. 214, pp. 109779

Abstract

Decarbonizing the hard-to-electrify transportation sector hinges on adopting renewable fuels. This transition to renewable fuels requires robust planning tools that can account for long-term demand uncertainty and evolving supply chain structures. This study develops a multi-stage stochastic mixed-integer linear program for the design of renewable fuel supply chain networks under uncertain demand. Unlike existing approaches that rely on deterministic or two-stage formulations, the proposed model captures sequential decision-making over time and enables adaptive responses to uncertainty. The model features a flexible multi-period, multi-stage framework supporting a wide range of products and feedstocks, considering seasonal and regional supply variability. A data-driven scenario generation and reduction framework is integrated to construct a tractable yet representative scenario tree. The model is applied to a European case study spanning 2021–2060, incorporating multiple fuel pathways and import strategies, including both long-term contracts and spot market procurement. The results shows the substantial economic advantage of our multi-stage approach. The stochastic model significantly outperforms deterministic planning, achieving a Relative Value of the Multi-Stage Stochastic Solution of 19.4% and an Expected Value of Perfect Information of 14.2%. Furthermore, the analysis demonstrates how the proposed model effectively balances strategic and tactical decisions under demand uncertainty. By accounting for sequential information updates, the framework reveals the interplay among technology costs, regional resource availability, and import strategies. These findings underscore the value of dynamic decision-making while offering practical guidance for policymakers and industry stakeholders in planning future renewable fuel supply chains. Overall, the proposed adaptive framework functions as a robust decision-support tool, enabling cost-efficient and flexible decarbonization planning.

Authors 3

1. Mina Farajiamiri corresponding Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University Aachen, NRW Germany

2. Jörn‐Christian Meyer Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University Aachen, NRW Germany

3. Grit Walther Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen University Aachen, NRW Germany