Research On Integrated Flexible Job-shop Scheduling With Multiple Process Plans By Ant Colony Optimization

Flexible job-shop scheduling problem is a complicated NP-hard combinational optimization problem.Along with diversified customer demands and increasingly fierce market competition,more and more enterprises adopt flexible production.Therefore,research on Flexible job-shop scheduling problem is helpful to improve production efficiency and reduce cost.Based on integration of process planning and scheduling,this paper considers flexible job-shop problem with operation and processing flexibility,known as FJSP with Multiple Process Plans(FJSP-MPP).Firstly,description methods may result in a combinational-explosive problem and oversize as the complexity of sequencing and processing flexibility of a job modelled increases.To solve this problem,a Four-Tuple based Mathematical Scheme,which can model a job with processing and/or operation flexibility well,is proposed.Secondly,based on the features of FJSP-MPP,an ACO-based approach is developed for solving FJSP-MPP.Without establishing upfront the disjunctive graph model for FJSP-MPP,the ant path oriented this problem is generated by FTMS.Due to the fact that the sequence of processed and the chosen processes both play an important role in satisfactory solution,an extra pheromone on node is introduced,and pheromone update strategy and transition probability rule are modified;To avoid sticking to stagnation situation,an adaptive evaporation coefficient mechanism that evaporation coefficient changes with best solution,is proposed.Besides,the global optimum solution preserved function is used.Thirdly,to validate this approach,four different scale computational experiments have been conducted.The results show that the proposed approach has a good optimization performance.Finally,the conclusions are summarized and some future research directions have been presented.