(392w) Decision Support for Optimal Green Hydrogen Distribution: A Mathematical Programming Model Considering Domestic Production and Imports

Green hydrogen is produced from water electrolysis powered by renewable electricity, making it a clean and sustainable energy carrier. Its ability to decarbonize hard-to-abate sectors, such as iron and steel, petrochemicals, fertilizer, and transportation, positions it as a critical solution for achieving net-zero emissions. A nation may satisfy its green hydrogen demand through imports, domestic production, or a mix of both. The decision to import, domestically produce, or adopt a mix of both depends on the availability of renewable energy resources, infrastructure capabilities, production costs, energy security considerations, and the nation's economic and environmental goals. The green hydrogen, whether imported, domestically produced, or both, is then distributed to end users from these facilities (domestic production and import terminals). Gaseous hydrogen is distributed through pipelines and tube trailers. In contrast, liquid hydrogen is distributed through tanker trucks and railcars. Due to the intermittent nature of renewable energy sources, the increasing number of end users, and their associated demands, the decision-making of green hydrogen distributors becomes challenging. Therefore, it is crucial to develop an optimal green hydrogen distribution strategy to ensure the demand satisfaction of end users and increase their reliability on the distributor. In this paper, we develop decision support for scheduling green hydrogen distribution from domestic production and import terminals to end users, minimizing the total cost.

Existing literature on decision support for green hydrogen supply chains primarily focuses on supply chain design. Garud et al.¹ presented a comprehensive optimization model for designing a green hydrogen production facility with a minimum landed cost of hydrogen. The proposed model was utilized to study green hydrogen production in various countries, including Saudi Arabia, Australia, Singapore, and Germany, focusing on the impact of geospatial solar irradiance on the facility design. Their findings suggested that hydrogen storage in tanks was more economical than storing renewable electricity in batteries (for green hydrogen production), and grid-connected facilities could have lower production costs but could not guarantee carbon-free hydrogen. Camelo et al.² proposed a mixed-integer linear programming (MILP) model to design a green hydrogen supply chain, focusing on the distribution of renewable energy and hydrogen from production plants to the end users. The MILP model aims to minimize total costs by determining the number, type, and location of production plants, transportation modes, and storage facilities. The model was demonstrated by a case study for Ceará, Brazil. The case study considered three scenarios with varying demands to assess cost reductions in hydrogen production. Their results indicated a 36.24% reduction in production costs between pessimistic and optimistic scenarios, offering valuable insights for decision-making in green hydrogen supply chain infrastructure and operations. Choi et al.³ proposed a MILP model for joint planning of an integrated supply chain of hydrogen and renewable energy (ISCHRE) system. The proposed model determined facility (production, storage, transportation, and transmission facilities) locations and capacities. The model also assisted in making operational decisions to meet renewable electricity and hydrogen demands while minimizing the total annualized cost. The model was used to design an optimal ISCHRE system for South Korea. The optimal ISCHRE system achieved a 20.8% reduction in annualized costs, increasing South Korea’s renewable energy penetration rate by 10%. Despite these scientific advancements, the existing literature does not consider the import of green hydrogen and the scheduling of transportation modes during distribution. We seek to address these research gaps through our model.

In this paper, we develop decision support for scheduling green hydrogen distribution. The problem considers a distributor which delivers green hydrogen from a set of domestic production and import terminals to end users spread over a geographical region. Each end-user has a known location and demand. Green hydrogen is distributed from these facilities (domestic production and import terminals) to end users through tanker trucks. One-way transportation time taken by tanker trucks to travel from the facilities to the end users and between the end users is known. Geographically close located end users are clustered together. As a result, the entire geographical region is divided into multiple clusters. A subset of tanker trucks is pre-allocated to each cluster, i.e., only these trucks can distribute hydrogen to the end users within that cluster. The tanker trucks pre-allocated to a given cluster can split between its end users. The number of split deliveries of a tanker truck in a trip is restricted to three. The distributor makes decisions related to tanker truck scheduling (loading and splitting) and volume distributed to end users over the planning horizon. The distributor aims to satisfy the demand of every end user, minimizing the total cost. The total cost is the sum of domestic production, transportation, and penalty (due to over-supply) costs. A mathematical programming model with the above-mentioned features and objective is developed. The proposed model is demonstrated by a case study having a planning horizon of one week. The case study is implemented in IBM ILOG CPLEX Optimization Studio v12.10.0. The computational results of the case study are also presented. We believe that the results from this paper would help develop critical insights regarding the development of decision support for scheduling green hydrogen distribution.

References

(1) Garud, S. S.; Tsang, F.; Karimi, I. A.; Farooq, S. Green Hydrogen from Solar Power for Decarbonization: What Will It Cost? Energy Convers. Manag. 2023, 286, 117059. https://doi.org/10.1016/j.enconman.2023.117059.

(2) Camelo, M. M.; De Andrade, C. F.; Prata, B. D. A. A Mixed-Integer Linear Programming Model for Optimizing Green Hydrogen Supply Chain Networks. Int. J. Hydrog. Energy 2025, 118, 134–145. https://doi.org/10.1016/j.ijhydene.2025.02.138.

(3) Choi, Y.; Kim, M.; Kim, S. H.; Heo, S. Synergy Evaluation for Joint Expansion Planning of Green Hydrogen and Renewable Electricity Supply Chains: A South Korea Case. Appl. Energy 2025, 381, 125123. https://doi.org/10.1016/j.apenergy.2024.125123.