基于并行协同的多车间协同调度问题研究

Multi-shop collaborative scheduling problem based on parallel collaboration

传统企业在实际生产中,其多个关联车间之间的生产计划与调度存在难以协作的问题。为此,针对多车间协同调度问题建立了调度模型,提出了一种多车间协同调度的并行协同进化遗传算法(PCE-GA),并且采用该算法对上述模型进行了求解。首先,以最小化订单完工时间为目标,建立了单目标调度模型;然后,采用了并行协同进化遗传算法,对上述单目标调度模型进行了求解,基于工件、机器、装配关系的三层整数编码的染色体编码方案,提出了一种协同适应度值计算的方法;最后,以某液压缸生产企业为例,针对单目标调度问题,采用该算法与单车间遗传算法(JSP-GA)、并行协同模拟退火算法(PCE-SA)分别进行了求解,并对其结果进行了比较,以验证PCE-GA算法的优越性。研究结果表明:采用PCE-GA算法得到的优化率为13.3%,比单车间作业调度遗传算法求解的数据优化11.5%,该结果证明了PCE-GA算法在解决多车间协同优化问题时的优越性。

In the actual production of traditional enterprises,there was a problem that it was difficult to coordinate the production planning and scheduling among its multiple associated workshops.Therefore,a scheduling model was established for the multi-shop collaborative scheduling problem,and a parallel co-evolutionary genetic algorithm(PCE-GA)for multi-shop collaborative scheduling was proposed,and the algorithm was used to solve the above model.Firstly,a single objective scheduling model was established to minimize the maximum order completion time,the parallel cooperative evolutionary genetic algorithm was applied to solve this optimal problem.Then,a chromosome coding scheme of three-layer integer coding based on workpiece,machine and assembly relationship was adopted in the algorithm.Using the mechanism of collaborative evolution,a method for calculating the value of cooperative fitness value was proposed.Finally,a hydraulic cylinder manufacturing enterprise was regarded as a study case.For the single-objective scheduling problem,the parallel cooperative evolutionary genetic algorithm was compared with the single-shop genetic algorithm and the parallel cooperative simulated annealing algorithm.The research results indicate that the optimization rate of the proposed parallel cooperative evolutionary genetic algorithm is 13.3%,which is 11.5%better than that of the single-shop genetic algorithm.The algorithm verifies the feasibility of the proposed model and the effectiveness of the proposed method.It is also concluded that the parallel cooperative evolutionary genetic algorithm performs better in solving the multiple shop collaborative optimization problem.