关键词: 资源动态调度, 机器故障率, 机器役龄, Pareto熵, 多目标粒子群算法, 并联加工系统

Abstract: Aiming at the parallel processing system,in order to deal with the dynamic scheduling problem of the processing resources caused by machine failure in the production process,considering the dynamic mutual restraint relationship between the system resource load and the failure rate,according to different sub-period division strategies,the processing time after the scheduling,Quantitative analysis of preventive maintenance time and system reliability,and constructing a multi-objective dynamic scheduling model.The Pareto entropy-based multi-objective particle swarm optimization algorithm is used to solve the model.By improving the individual optimal solution selection strategy,the diversity of the optimal solution selection is improved.The evolution of the population is estimated by the difference entropy,and the search ability of the algorithm is improved.Taking an enterprise's processing system as an example,this model is used to design an integrated scheduling scheme for processing resources.Through the multi-objective weights obtained by fuzzy decision,the optimal scheduling scheme is selected,and several parameters are compared to verify that in the scheduling process of processing resources,The integration considers the dynamic machine failure rate and the feasibility of dividing the sub-periods based on the maximum working-age constraints of the machine,not only can deliver on-time delivery of orders,but also can make enterprises have better flexibility to deal with emergency orders that may come.

Key words: resource dynamic scheduling, machine failure rate, machine working-age, Pareto entropy, multi-objective particle swarm optimization algorithm, parallel machining system