Solving the blocking job-shop scheduling problem considering transportation and the no buffer constraint: a performance and price-based auction approach

This paper discusses the blocking job-shop problem (BJSP) under the transportation and no buffer constraint. Different from the classical job-shop problem (JSP), the BJSP under the transportation constraint occurs after a job is processed on a machine, which is then unable to process subsequent jobs until the first job is transferred to another machine by an automated-guided vehicle (AGV), and describe it as a mixed-integer linear programming model. To solve the problem, a performance and price-based auction approach is designed, it can obtain feasible solutions with high quality, and then, computational efficiency of the neighbourhood structure in the local search is improved significantly over that in prior studies. The restoration strategy for infeasible solutions is also extended. In a series of experiments, our proposed approach identifies better solutions to the problem compared with prior methods in the extant research. In the comparison, the maximum improvement in objective function is more than 70%. According to the experimental results using SPSS, the proposed approach proves to be efficient and stable, especially suitable for solving large-scale numerical examples.