Abstract
In this paper, we present a novel variation of the network design problem to obtain a minimum setup and operational cost hydrogen refuelling infrastructure. The problem is modelled as a mixed integer linear programming problem which incorporates both location and supply considerations. More specifically, the model is employed to determine the optimal locations for hydrogen refuelling stations and the methods of supplying and delivering hydrogen to each station. The supply of hydrogen is considered in two forms, on-site or localised supply, and off-site or centralised supply. In the latter case, hydrogen is distributed to the refuelling stations by means of either tube trailers or a hydrogen pipeline. The complexity of solving the network design problem primarily arises from the routing decisions for the hydrogen pipeline. To address this, we derived two different formulations and developed a branch-and-cut algorithm. We then compared the two formulations in a preliminary computational study where we found that the generalised cut-set formulation paired with a polynomial time separation procedure was the most efficient. Finally, we applied our model to a real-world case study in the north of England. The results demonstrated that incorporating hydrogen delivery by pipelines significantly improved the objective function.
Original language | English |
---|---|
Article number | 2509704 |
Journal | Journal of the Operational Research Society |
Early online date | 2 Jun 2025 |
DOIs | |
Publication status | E-pub ahead of print - 2 Jun 2025 |
Keywords
- Discrete optimisation
- hydrogen refuelling infrastructure
- location science
- exact solution approach