Modelling and optimization of distributed assembly hybrid flowshop scheduling problem with transportation resource scheduling

Abstract

In the distributed assembly flowshop scheduling problem (DAFSP), the transportation resources between the production and assembly stages are rarely scheduled. However, for production managers, the transportation stage is important as it directly influences the cooperation between the production and assembly phases. Hence, in this study, a critical variant of the DAFSP, called the distributed assembly hybrid flowshop scheduling problem with transportation resource scheduling (DAHFSPT), is evaluated. DAHFSPT considers several vehicles for component delivery, and the early assembly is allowed in the assembly site. A mixed-integer linear programming model of the problem is developed to minimize the makespan and total energy consumption simultaneously. To solve the problem, an improved multi-objective evolutionary algorithm based on decomposition (IMOEA/D) is proposed. In the IMOEA/D, a well-designed encoding method, and eight heuristics decoding methods are embedded into the initialization operator to generate a high-quality initial population. A problem-based local search operator is presented to enhance the performance of the algorithm and accelerate its convergence. In addition, an elimination-reinitialization operator is designed to avoid optimal local solutions. A total of 200 instances based on classical benchmarks are formulated to verify the effectiveness of each improvement component in the IMOEA/D. Comprehensive experiments indicate that the IMOEA/D outperforms five well-known multi-objective algorithms concerning solution quality and distribution in solving the DAHFSPT.