New Formulation and Solution in PCB Assembly Systems with Parallel Batch processors

This paper considers the scheduling problem of parallel batch processing machines with nonidentical job size and processing time. In this paper, a new mathematical model with ready time and batch size constraints is presented to formulate the problem mathematically, in which simultaneous reduction of the makespan and earliness-tardiness is the objective function. In recent years, the nature-inspired computational intelligent algorithms have been successfully employed to achieve the optimum design of different structures. Since the proposed model is NP-hard, a metaheuristic algorithm according to a harmony search algorithm is developed and analyzed for solving the batch processing machine scheduling problem addressed in the current paper. Various parameters and operators of the proposed harmony search algorithm are discussed and calibrated by means of the Taguchi statistical technique. In order to evaluate the proposed algorithm, instance problems in concordance with previous research are generated. The proposed algorithm and basic harmony search, improved harmony search and global best harmony search are solved and the results of all the algorithms are compared. The conclusion reveals that the proposed algorithm performs better than the other algorithms.