Abstract: To make the product processing and assembling in parallel and to improve production efficiency,the comprehensive scheduling for two purposes were processed,and a scheduling algorithm based on virtual component level division coding was proposed.The model of product comprehensive scheduling problem was given,and the genetic algorithm was adopted to solve this problem.A novel division coding strategy which ensured all initial solutions feasible was designed based on rank of virtual components,which reflected the sequence constraints of parts processing and assembly.Feasible crossover and mutation methods were given for division coding to ensure the feasibility of offspring individuals and avoid the check and repair work.The decoding method was presented for parts processing and assembly constraints.The designed genetic algorithm was tested by literature example.Experimental results showed that the proposed algorithm had good solution speed and quality,and its validity was verified.In addition,the proposed encoding,crossover and mutation methods provided valuable references for other intelligent algorithms to solve this problem.

Key words: product comprehensive scheduling, virtual component, division coding, genetic algorithms

摘要： 为使产品加工与装配并行进行、提高生产效率,将二者一同处理并进行综合调度,提出一种基于虚拟零部件级别分区编码的调度算法。给出了产品综合调度的问题模型,采用遗传算法对问题进行求解。该算法设计了一种新颖的基于虚拟零部件级别的分区编码方式,体现了零部件加工装配的顺序约束,保证了所有初始解的可行性;研究了针对分区编码可行的交叉和变异操作方式,保证了所有子代个体的可行性,避免了检测修复等工作;给出了具有零部件加工装配顺序约束的解码方法。采用文献中已有实例对算法进行了测试,对比实验结果表明,该算法具有良好的求解速度和质量,验证了其有效性;所设计的编码方式和交叉变异操作为其他智能算法求解该问题提供了借鉴价值。

关键词: 产品综合调度, 虚拟零部件, 分区编码, 遗传算法