中文摘要：

目前,印刷电路板（PCB）组装业面临着多品种小批量的生产环境.提高生产效率,实现组装过程的最优化调度是PCB组装优化研究的核心问题,也是电子制造企业关注的重点.随着新的生产技术——整体切换台车的出现,吸嘴切换在多品种PCB组装中成为新的瓶颈.本文考虑吸嘴切换,对多品种PCB组装生产时间进行优化,建立了整数规划模型,并设计混合算法优化PCB组装的总完工时间.首先,采用遗传算法对多品种PCB进行分组;其次,对组内吸嘴分配问题采用改进的解决席位公平分配问题的增量算法,使得组内生产时间最短;最后,对组间吸嘴切换问题采用改进的KTNS（keep tool needed soonest）算法,使得切换次数最少,并通过实例分析证明了算法的有效性.

英文摘要：

At present,the characteristics of printed circuit board（ PCB） assembly is high-mix and low-volume. As a result,to enhance productivity and optimize PCB assembly efficiency become an important problem. With the invention of interchangeable feeder trolley,the nozzle setup becomes the new bottleneck of PCB assembly for multiple PCB types. An integer programming model of multiple PCB types assembly was set up in consideration of nozzle setups,and a hybrid algorithm for minimizing the total makespan was proposed.Firstly,the PCB grouping problem was solved by genetic algorithm. Then,the head allocation problem was solved by modifying increment method which is used to solve apportionment problem and minimize the processing time in the same group.Finally,the nozzle switching problem between different groups was solved by modified keep tool needed soonest（ KTNS） algorithm. Numerical experiments were carried out to examine the performance of the proposed algorithm.