Stochastic optimization for strategic planning of efficient and sustainable hydrogen supply chain networks under demand uncertainty

Hydrogen has emerged as a crucial energy carrier, playing a vital role in the global transition towards a low-carbon economy. Despite its growing strategic importance, accurately forecasting hydrogen demand remains challenging due to the influence of diverse and interdependent factors. An effective and sustainable infrastructure planning requires modeling frameworks that explicitly incorporate demand uncertainty inherent in hydrogen market projections. This study proposes a stochastic mixed-integer linear programming (SMILP) framework for optimizing the strategic development of hydrogen supply chain (HSC) infrastructure, accounting for the demand uncertainty. Demand uncertainty is modeled using discrete probabilistic realizations across multiple time periods. A novel approach is presented in incorporating and quantifying the expected penalty costs for unfulfilled demand through a linear formulation. The proposed model aims to minimize the total system cost, encompassing both economic and emission costs across all stages of the HSC, including production, conditioning and storage, transportation, and re-conditioning. This study uses Qatar as an illustrative case to evaluate three distinct investment strategies to test the model's applicability in real-world scenarios. Across the different strategies, the end-to-end levelized cost of hydrogen (LCOH) for the entire supply chain ranges from $3.55/kg to $3.68/kg, while the production LCOH ranges from $1.18/kg to $1.23/kg. The findings highlight the importance of adopting a balanced approach to infrastructure planning, particularly under volatile and fluctuating market conditions.