A genetic algorithm-based matheuristic for hydrogen supply chain network problem with two transportation modes and replenishment cycles. (January 2019)

• Record Type: Journal Article
• Authors: Woo, Young-Bin; Kim, Byung Soo
• Is Part Of: Computers & industrial engineering. Volume 127(2019)
• Page Start: 981
• Page End: 997
• Publication Date: 2019-01
• Subjects: Hydrogen supply chain network -- Matheuristic -- Genetic algorithm -- Mathematical programming -- Replenishment cycle; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2018.11.027
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: We address for hydrogen supply chain network (HSCN) problem with multi-transportation modes and replenishment cycles. A mixed integer nonlinear programming model (MINLP) is developed to solve HSCN problem. A novel genetic algorithm-based matheuristic (GAM) is proposed to solve HSCN problem. An application study for HSCN at Jeju Island, South Korea is conducted by using GAM. Abstract: This paper addresses a hydrogen supply chain network problem (HSCNP) with two transportation modes and replenishment cycles. For determining an optimal hydrogen supply chain network (HSCN), a mixed integer non-linear programming (MINLP) model is developed. Due to the intractability caused by non-linear terms of the MINLP model, a genetic algorithm-based matheuristic (GAM) is proposed by including a mixed integer linear programming (MILP) combined within the procedure of genetic algorithm (GA). The performance of GAM is compared with two GAs by using randomly generated instances. Then, several sensitivity analyses of demand, transportation cost, and inventory carrying cost are conducted to evaluate an effect on a configuration of HSCN. Finally, optimal network configurations and selection of transportation modes are estimated under different future demand scenarios of Jeju Island, South Korea.