Optimizing renewable fuel supply chain network design under sustainability and uncertainty considerations

RWTH Publications (RWTH Aachen)

Abstract

This dissertation develops an integrated decision-support framework for the design of renewable fuel supply chains that support decarbonization of hard-to-electrify transport sectors under stringent sustainability constraints and pervasive uncertainty. Motivated by the need to align large-scale fuel production with climate targets, limited land and water resources, and existing uncertainties, the work addresses four overarching research questions: (i) can eco-nomic, land use, and water use trade-offs in EU-wide renewable fuel supply chains be quantified and optimized; (ii) how does variability in renewable electricity sources affect optimal infrastructure and import strategies; (iii) how can advanced uncertainty-aware planning approaches support adaptive, time-dependent decision-making along evolving fuel-demand trajectories; and (iv) what combined insights emerge from integrating sustainability and uncertainty modeling for EU-level planning and policy. Methodologically, the dissertation builds a flexible strategic–tactical mixed-integer linear programming framework based on a Resource–Technology–Network representation for multi-period, multi-feedstock, multi-product renewable fuel supply chains producing Fischer–Tropsch fuels, methanol, hydrogen, and ammonia from biomass, CO₂, and renewable electricity. Environmental sustainability is captured by extending the economic objective with explicit land use and water use objectives, which are jointly optimized using the augmented ε-constraint method to generate Pareto-efficient frontiers. Tactical uncertainty in renewable electricity availability is represented through a two-stage stochastic programming model that links long-term investment decisions with tactical recourse and distinguishes between contracted and spot imports; renewable electricity variability is captured via scenario trees derived from multi-decadal capacity factor data. Strategic, time-dependent demand uncertainty is addressed through a multi-stage stochastic programming model that employs autoregressive demand processes and forward scenario reduction to construct tractable scenario trees, enabling sequential infrastructure adaptation under non-anticipativity constraints. All models are applied to an EU+UK case study for transport fuel demand over 2021–2060.The results demonstrate that no single renewable fuel supply chain configuration simultaneously minimizes cost, land, and water use, but that substantial environmental improvements can be achieved at moderate cost premiums. The multi-objective analysis reveals dense Pareto fronts and a compromise solution in which large reductions in land and water requirements are attainable with relatively small increases in total system cost, while cost-optimal pathways are dominated by renewable electricity–based technologies and rely structurally on imports. The two-stage stochastic programming results show that explicitly modeling renewable electricity variability induces more flexible yet robust infrastructure portfolios, with strong regional specialization in solar and wind deployment, late but significant uptake of contracted imports, and measurable economic benefits compared with deterministic planning. The multi-stage stochastic programming analysis highlights adaptive investment paths that hedge through early diversification and converge as information is revealed, quantifying the economic penalty of demand uncertainty and the cost savings from multi-stage planning relative to static designs. Together, the three papers provide a coherent, quantitative basis for designing renewable fuel supply chains that jointly balance economic efficiency, land and water constraints, and robustness to both supply-side and demand-side uncertainties, yielding policy-relevant insights on technology portfolios, infrastructure timing, and the long-term role of imports in EU renewable fuel strategies.

Authors 1

1. Mina Farajiamiri corresponding Aachen

   RWTH Aachen University

   Affiliation as printed

   RWTH Aachen