管道工具喷粉线生产调度建模与算法  

Research on modeling and algorithms for the scheduling problem of pipe tools powder painting production

作　　者：胡章勇[1] 苏强[2] 霍佳震[2] 

机构地区：[1]中国商用飞机有限责任公司上海飞机设计研究院,上海200232 [2]同济大学经济与管理学院管理科学与工程系,上海200092

出　　处：《系统工程理论与实践》2011年第12期2348-2354,共7页Systems Engineering-Theory & Practice

基　　金：国家自然科学基金(70832005;71090404;71072026)

摘　　要：针对不同品种、不同材质和不同颜色管道工具喷粉生产调度问题,以生产成本最小化为优化目标,研究其优化调度方法.首先,建立这个问题的混合整数非线性规划模型(MINLP);其次,针对该问题设计出相应的贪婪随机自适应搜索算法(GRASP)和遗传算法(GA);在此基础上,提出两种算法相集成的GRASP+GA算法.应用生产实例数据分析比较了这三种算法,验证了GRASP+GA算法的优越性.In this paper, the pipeline tool powder painting scheduling problem with different products, different materials and different colors, is studied to minimize the production cost. First, a mixed integer nonlinear programming model （MINLP） is constructed for this problem. Then two algorithms of the greedy randomized adaptive search procedure （GRASP） and genetic algorithm（GA） are designed. Furthermore, an integrative algorithm GRASP＋GA is constructed for this problem. With the real production data the three algorithms are tested and the calculation results demonstrate that the GRASP＋GA algorithm is much better than GRASP or GA.

关 键 词：喷粉线 生产调度 贪婪随机自适应搜索算法(GRASP) 遗传算法(GA) GRASP+GA算法 

分 类 号：F406[经济管理—产业经济] TP301[自动化与计算机技术—计算机系统结构]