面向蚁群算法的车间调度问题研究-计算机应用技术专业毕业论文.docxVIP

哈尔滨娌工大学T学坝i：学位论文面向蚁群算法的车间 哈尔滨娌工大学T学坝i：学位论文 面向蚁群算法的车间 调度问题研究 摘要 车间调度是一个典型的NP难问题，理论上已经证明想在多项式时问内 对这‘类问题找到全局最优解是不可能的。蚁群算法是近年来兴起的‘种寻 优算法，特别在解决组合优化问题中被越来越多的人所采用。本文应用蚁群 算法求解两个车间调度问题，提出建立蚂蚁初始点的概率优先权分布方法， 利用蚁群算法的正反馈和并行搜索特点提高解的质量和稳定性。 1．通过对基本蚁群算法的实现过程和Job—shop调度问题的分析和研 究，给出了求解Job．shop调度问题的析取图模型，并给出了详细的ACA蚁 群算法求解Job．shop调度问题的算法描述，然后通过对28个Benchmark实 例进行模拟。试验结果表明：蚁群算法可以有效地求解车间调度问题：对于 个别实例可以达到最优解：蚁群算法的平均性能优于遗传算法。 2．通过对摹本蚁群算法的实现过程和流水调度问题进行研究，给出了 求解流水调度问题的解构造图，提出了两种蚁群算法来求解以总流程时川最 小为目标的流水调度问题一FACA蚁群算法和PAcA蚁群算法，分别给出了 FACA蚁群算法和PACA蚁群算法的参数初始化方式、相对应的信息素更新 方法和概率分布规则，并提出了局部搜索模式，最后进行模拟试验，对这两 种算法进行比较，试验结果表明：对于较大规模的问题，PACA蚁群算法要 优于FACA蚁群算法，而对于较小规模的问题，FACA蚁群算法优于PACA 蚁群算法。平均来说，PACA蚁群算法的性能优于FACA蚁群算法。 关键词车间调度；makespan；蚁群算法；启发式算法 哈尔滨埋1．大学『：学倾l。学位论文RESEARCH 哈尔滨埋1．大学『：学倾l。学位论文 RESEARCH oN AN ANT CoLoNY ALGoRITHM FOR JoB SHoP SCHEDULING PRoBLEM Abstract The Job—shop scheduling problem is a NP-hard The fact that it is impossible to find the global optimum in polynomial complexity has been proved Ant colony algorithm(ACA)，which is an optimization searching technique，has been widely noticed during recent years，An ant colony algorithm(ACA for short)is proposed for two Job-shop scheduling problem．A probability priority is introduced for the initial distribution of ants．Moreover，quality and robustness of solutions are improved by the nature of feedback and parallel paradigm ofACA 1．Through the analysis and studying of Ant colony algorithm and Job—shop scheduling problem，a disjunctive graph is introduced．The detail description of the proposed ACA algorithm is described．Last，ACA and GA are run on 28 instances with different sizes．Experimental results show that ACA can efficiently solve Job-shop scheduling problem and can obtain the optimums on some instances As well，ACA outperforms GA in performance on average． 2．Through the analysis and studying of Ant colony algorithm and Flow shop scheduling problem，a construction graph is introduced．Two ant colony algorithms(FACA and PACA)are proposed and analyzed for s