Job-shop cell-scheduling problem with inter-cell moves and automated guided vehicles

Abstract

This paper addresses part scheduling problems (CPS problems) in the context of the need for exceptional parts to visit machines among cells and to be transferred via an automated guided vehicle (AGV) in order to minimize the process make-span. A nonlinear mathematical programming model is proposed to determine the sequences of the parts processed on the machines, which is solved by a two-stage heuristic algorithm. Because of the complexity of the CPS problem, it is divided into two sub-problems: the Intra-cell part scheduling problem (Intra-CPS) and the Inter-CPS. The two-stage heuristic algorithm consists of a local search combined genetic algorithm (LSC-GA) whose neighborhood is based on the disjunctive graph model and a heuristic algorithm for finding a lower make-span through roulette selection (LMW-HEU). When solving the Intra-CPS problem, the sequence of parts processed by each machine and the sequence of parts to be transferred by AGV within manufacturing cells are determined cell by cell by the LSC-GA. Then, the sequence of parts from different cells that need to be processed on the same machine is determined by LMW-HEU in the Inter-CPS problem. A series of experiments were conducted to test the two-stage algorithm. The results demonstrate the effectiveness of the proposed two-stage algorithm.