混合遗传算法在Job—shop调度问题中的应用

首先建了Ｊｏｂ－ｓｈｏｐ调度问题的神经网络模型，根据这种模垢特点。提出了求解复杂Ｊｏｂ－ｓｈｏｐ调度问题的混合遗传算法。     

Job- shop 提前/拖期调度问题的研究

基于模糊控制和遗传算法,提出了求解Ｊｏｂ－ｓｈｏｐ提前／拖期间问题的联合算法,用遗传算法确定可行调度序列,然后用模糊控制器对开工时间加以调整,模糊控制的引入为有效地求解Ｊｏｂ－ｓｈｏｐ提前／拖期调度总理２提供了新的方法,仿真实验证明了联合自救的有效性。     

混合遗传算法在Job-shop调度问题中的应用

首先建立了Ｊｏｂ－ｓｈｏｐ调度问题的神经网络模型，根据这种模型的特点，提出了求解复杂Ｊｏｂ－ｓｈｏｐ调度问题的混合遗传算法．仿真结果表明了本文方法的有效性，在运行时间和最优率方面具有较好的优势．     

解Job－shop调度问题的神经网络方法

研究用神经网络方法解决Ｊｏｂ－ｓｈｏｐ调度问题．首先描述解Ｊｏｂ－ｓｈｏｐ调度问题的算法，然后给出这一算法及其网络性质的理论结果．仿真实验结果证明了该方法是可行的．最后，针对几类典型调度问题的解决进一步说明了这一方法的优势．     

基于遗传算法的滚动调度策略*

本文研究了动态加工环境下的一类Ｊｏｂ－Ｓｈｏｐ调度问题，提出了一种基于遗传算法的滚动调度策略，其要点是：１）借鉴预测控制的思想，采用ｔｉｍｅ－ｂａｓｅｄ和ｊｏｂ－ｂａｓｅｄ的滚动调度策略适应动态环境和要求的多变性。２）以遗传算法和分派规则相结合，处理考虑与操作序列有关的工件安装时间和工件到期时间约束的复杂调度问题。文中给出了在工件到期时间发生改变的动态环境中两种滚动调度算法的调度结果，并与静态调度     

用遗传算法与自适应神经网络混合方法解Job—shop调度问题

提出一种用遗传算法结合基于约束满足的自适应神经网络进行Ｊｏｂ－ｓｈｏｐ调度问题求解的混合方法。遗传算法被用来进行迭代寻优。当前代经交叉和变异后生成的染色体对应非可行解，由自适应神经网络运算后得到可行解，对应的染色体作为新一代染色体。仿真表明该算法是快速有效的。     

Job—shop多机实时调度的并行算法

本文针对ＭＩＭＤ并行机对一般的Ｊｏｂ－ｓｈｏｐ调度提出实时调度的并行算法，通过分析复杂性和加速比以及实例，说明并行算法对求大批工件多台机器加工的最优调度的优越性。     

用遗传算法与自适应神经网络混合方法解Job-shop调度问题

提出一种用遗传算法结合基于约束满足的自适应神经网络进行Ｊｏｂ—ｓｈｏｐ调度问题求解的混合方法。遗传算法被用来进行迭代寻优。当前代经交叉和变异后生成的染色体对应非可行解，由自适应神经网络运算后得到可行解，对应的染色体作为新一代染色体。仿真表明该算法是快速有效的     

车间作业调度遗传算法中的交叉算子研究

针对车间作业调度遗传算法，在车间作业调度数学表达模型基础上，讨论车间作业调度遗传算法交叉算子的宏观设计与微观设计，提出了ＪＳＳ（ＪｏｂＳｈｏｐＳｃｈｅｄｕｌｉｎｇ）交叉算子设计原则，并给出其应用结论，说明了交叉算子设计的有效性。     

解Job-shop调度问题的混合模拟退火进化规划

提出运用混合模拟退火进化规划（ＳＡＥＰ）求解Ｊｏｂ－ｓｈｏｐ调度问题．首先介绍了ＳＡＥＰ和进化规划（ＥＰ）的不同选择方法以及他们的变异算子,最后给出了仿真实例,并比较了这两种算法的优劣     

Minimizing sum of the due date assignment costs,maximum tardiness and distribution costs in a supply chain scheduling problem

In production systems, manufacturers face important decisions that affect system profit. In this paper, three of these decisions are modelled simultaneously: due date assignment, production scheduling, and outbound distribution scheduling. These three decisions are made in the sales, production planning and transportation departments. Recently, many researchers have devoted attention to the problem of integrating due date assignment, production scheduling and outbound distribution scheduling. In the present paper, the problems of minimizing costs associated with maximum tardiness, due date assignment and delivery for a single machine are considered. Mixed Integer Non-Linear Programming (MINLP) and a Mixed Integer Programming (MIP) are used for the solution. This problem is NP-hard, so two meta-heuristic algorithms, an Adaptive Genetic Algorithm (AGA) and a Parallel Simulated Annealing algorithm (PSA), are used for solution of large-scale instances. The present paper is the first time that crossover and mutation operators in AGA and neighbourhood generation in PSA have been used in the structure of optimal solutions. We used the Taguchi method to set the parameters, design of experiments (DOE) to generate experiments, and analysis of variance, the Friedman, Aligned Friedman, and Quade tests to analyse the results. Also, the robustness of the algorithms was addressed. The computational results showed that AGA performed better than PSA.     

解并行机模糊调度问题的自适应遗传算法

针对家纺企业车间调度的实际情况,建立了一种产品优先级约束的模糊车间调度模型。在模型中,完工时间和交货期都是模糊的,交货期平均满意度最大为调度目标。基于此模型,提出了一种自适应的遗传算法,该算法通过比例选择及局部搜索保证种群的优良特性,并通过自动调节变异率和交叉率的方式保证种群的多样性,有效跳出局部收敛。仿真结果表明,自适应遗传算法能有效求解,并优于免疫遗传算法。     

An asynchronous genetic local search algorithm for the permutation flowshop scheduling problem with total flowtime minimization

In this study, the permutation flowshop scheduling problem with the total flowtime criterion is considered. An asynchronous genetic local search algorithm (AGA) is proposed to deal with this problem. The AGA consists of three phases. In the first phase, an individual in the initial population is yielded by an effective constructive heuristic and the others are randomly generated, while in the second phase all pairs of individuals perform the asynchronous evolution (AE) where an enhanced variable neighborhood search (E-VNS) as well as a simple crossover operator is used. A restart mechanism is applied in the last phase. Our experimental results show that the algorithm proposed outperforms several state-of-the-art methods and two recently proposed meta-heuristics in both solution quality and computation time. Moreover, for 120 benchmark instances, AGA obtains 118 best solutions reported in the literature and 83 of which are newly improved.     

An approaching genetic algorithm for automatic beam angle selection in IMRT planning

A method named approaching genetic algorithm (AGA) is introduced to automatically select the beam angles for intensity-modulated radiotherapy (IMRT) planning. In AGA, the best individual of the current population is found at first, and the rest of the normal individuals approach the current best one according to some specially designed rules. In the course of approaching, some better individuals may be obtained. Then, the current best individual is updated to try to approach the real best one. The approaching and updating operations of AGA replace the selection, crossover and mutation operations of the genetic algorithm (GA) completely. Using the specially designed updating strategies, AGA can recover the varieties of the population to a certain extent and retain the powerful ability of evolution, compared to GA. The beam angles are selected using AGA, followed by a beam intensity map optimization using conjugate gradient (CG). A simulated case and a clinical case with nasopharynx cancer are employed to demonstrate the feasibility of AGA. For the case investigated, AGA was feasible for the beam angle optimization (BAO) problem in IMRT planning and converged faster than GA.     

基于改进蚁群算法的关键链项目进度管理研究

项目进度管理是项目管理工作中的重要内容,关键链法是目前项目管理中较为常用的进度管理方法之一,其本质为多约束优化问题。结合混沌运动与遗传算法的优点,对蚁群算法进行改进,并将其应用于解决关键链项目管理的优化调度问题。克服了蚁群算法由于前期信息素匮乏而导致的需要较长时间进行搜索、容易得到局部最优解的缺点,使混合算法的搜索范围有所增加,蚁群群体的进化速度得到提升,并保持了蚁群算法鲁棒性及收敛性,且算法的计算精度较高,求解速度较快。实例对比分析表明,在求解关键链项目进度管理问题上,混沌蚁群进化算法比遗传蚁群算法更具有优势。     

基于自适应遗传算法的路径测试数据生成

针对简单遗传算法容易产生早熟收敛的问题,提出一种自适应遗传算法,用以自动生成测试数据。通过把程序插装法与该遗传算法相结合,实现了路径测试数据的自动生成。将三角形分类程序作为实例对其进行性能测试,实验结果表明,基于自适应遗传算法的测试数据自动生成系统能自动改变选择概率和交叉概率,提高了自动生成测试数据的效率。     

一种新的图象分割自适应算法的研究

首先对自适应遗传算法的变异算子进行了改进,对单点变异算子与双点变异算子的结合能有效地改善局部收敛进行了验证,然后提出了一种新的用自适应遗传算法分割图象的方法,并与传统的Ｏｔｓｕ方法、灰度差直方图法和基于熵的方法作了比较。     

A flexible tolerance genetic algorithm for optimal problems with nonlinear equality constraints

A hybrid method called a flexible tolerance genetic algorithm (FTGA) is proposed in this paper to solve nonlinear, multimodal and multi-constraint optimization problems. This method provides a new hybrid strategy that organically merges a flexible tolerance method (FTM) into an adaptive genetic algorithm (AGA). AGA is to generate an initial population and locate the “best” individual. FTM, serving as one of the AGA operators, exploits the promising neighborhood individual by a search mechanism and minimizes a constraint violation of an objective function by a flexible tolerance criterion for near-feasible points. To evaluate the efficiency of the hybrid method, we apply FTGA to optimize four complex functions subject to nonlinear inequality and/or equality constraints, and compare these results with the results supplied by AGA. Numerical experiments indicate that FTGA can efficiently and reliably achieve more accurate global optima of complex, nonlinear, high-dimension and multimodal optimization problems subject to nonlinear constraints. Finally, FTGA is successfully implemented for the optimization design of a crank-toggle mechanism, which demonstrates that FTGA is applicable to solve real-world problems.     

作业车间环境下的采购-生产-配送联合调度模型

针对生产-配送联合调度（IPDS）模型较少考虑复杂生产环境以及采购环节的问题,建立了在作业车间环境下,以最小化订单完成时间为目标的采购-生产-配送联合调度（IPPDS）模型,并采用改进的动态人工蜂群（DABC）算法进行求解。根据IPPDS问题的特征,首先,采用二维实数矩阵的编码方式,实现任务（加工与运输）与资源（设备与车辆）的匹配关系;其次,采用基于工艺过程的解码方式,并在解码过程中针对不同任务设计了满足约束条件的方法,来保证解码方案的可行性;最后,在算法过程中设计了引领蜂与跟随蜂的动态协调机制和局部启发式信息。通过实验给出DABC适当的参数区间,对比实验结果表明,IPPDS策略相较于分段调度和IPDS策略,调度时间分别缩短了35.59%和30.95%;DABC相较于人工蜂群（ABC）算法求解效果平均提升了2.54%,相对于改进的遗传算法（AGA）求解效果平均提升了6.99%。因此,IPPDS策略能更快速地满足客户需求,而DABC算法既减少需设置的参数,又具有良好的探索和开发能力。     

遗传算法中自适应方法的比较和分析

分析了前人提出的具有代表性的自适应遗传算法,使用23个测试函数对SGA和3种AGA进行实验比较,讨论并总结出各种AGA的优劣所在,为新研究理念的提出提供基础,也为工业应用提供一个参考标准.实验结果表明,基于聚类分析的AGA在算法性能上较其它自适应遗传算法更优,具有很高的实用价值和发展前景.