Abstract: The integrated optimization problems for mixed model shop, which composed of a kind of coordinated scheduling for job shop and flow shop with assembly are investigated. Multi-objectives are considered simultaneously: minimizing the make-span in the job shop and minimizing the total make-span in assembly line, in order to ensure the inventory of work piece in process of buffers and reduce the waiting time in assembly line, consider the penalty for the earlier completion in fabrication line and wait in assembly line. The multi-objective integrated of mixed shop model is presented. It takes advantage of the parallelism between fabrication and assembly to determine the sequence of fabrication processing and assembly and cuts down the make-span of production, reduces the processing cost and improves the rate of equipment utilization. The multi-objective GA is designed for solving the scheduling problem, in which the two-section real number encoding method is put forward, the select, crossover, mutation operations and the elitist strategy are designed to ensure the priority in processing and avoid the effect of infeasible solution. And finally, an example from freezer production is given to test the model and algorithms presented herein. The results show that the model is effective and excellent.

Key words: fabrication and assembly, integrated scheduling, multi-objective optimizing, Genetic Algorithm（GA）

摘要： 为解决一类具有多品种混流加工作业车间和流水装配车间的两阶段集成调度优化问题，以加工线最大完工时间和产品总生产完工时间最小为目标，并考虑通过对零部件加工提前完工和装配线等待施加惩罚系数，以保证缓冲区在制品库存和装配过程均匀连续生产，建立加工与装配车间集成调度的多目标优化模型，充分利用加工和装配工序之间存在的并行性，合理确定零部件加工顺序和装配排序，以缩短产品生产周期，降低生产成本，提高生产设备利用率；同时针对所建立的模型，设计遗传算法进行求解，采用零件加工和产品装配的两段实数编码，以稳态复制对群体进行选择，对交叉和变异算子进行设计，以保证新个体满足工序先后约束的可行性，避免了非可行解的混入影响优化结果；最后通过实例验证所建调度模型的可行性和算法的有效性。

关键词: 加工与装配, 集成调度, 多目标优化, 遗传算法