求解异构并行机调度问题的混合多目标教-学优化算法

以异构并行机调度问题为研究对象,考虑了一类以优化总加权完工时间和加权延误总和的调度问题。首先,基于问题描述构建了该问题的混合整数规划模型。其次,提出了混合多目标教-学优化算法。在算法设计中,结合问题的特点设计序列编码方法,并采用分解技术来实现多目标调度问题的求解。此外,该算法通过融合多种交叉算子来定义个体进化过程,并通过与变邻域搜索算法的混合来提升其优化效果。最后,给出了仿真实验与分析,测试结果验证了多目标教-学优化算法求解该调度问题的优越性。     

带运输时间混合流水车间成组调度的协同进化文化基因算法EI北大核心CSCD

研究了一类带有序列相关准备时间和阶段间运输时间的混合流水车间成组调度问题,以最小化最大完工时间为目标建立混合整数线性规划模型,结合问题特征提出一种协同进化文化基因算法.算法采用置换序列的方式对工件组间调度、各工件组内工件间调度以及各工件组在各阶段上并行机的指派3个子问题进行统一编码,基于负载均衡思想和改进的先到先得策略将染色体解码为问题的可行解;进化过程中采用多种遗传算子执行全域搜索,并设计了一种基于破坏和重新构造的协同进化局部搜索策略.通过不同问题规模的数据实验和与对比算法的比较分析,验证了所提模型和算法的有效性.     

混合并行机调度问题的多目标优化模型及算法

针对生产工序的合并造成一种串并联共存的生产布局,研究了一种特殊的混合并行机调度问题,并考虑以最小化总流水时间和最小化总延迟工件数量为目标的多目标调度问题,建立了混合整数规划模型.针对模型特点,设计了一种改进的非支配排序遗传算法进行求解,采用基于启发式方法的初始种群生成方式以提高种群的质量和多样性,并引入一种局域搜索策略以改善求解算法所获得的非支配解的质量及分布性.通过对大量数值算例进行仿真实验,并与典型的多目标优化算法进行比较,结果表明所提出的模型和算法在收敛性、分布性及极端点质量方面均具有优势,能够较好的解决多目标混合并行机调度问题.     

求解考虑机器调整时间的并行机分批优化调度问题

基于目前车间调度问题是以单个或整批进行生产加工的并行机调度模型已不再符合实际工况下的车间生产。提出以最小化最大完工时间为优化目标,对遗传差分进化混合算法,灰狼差分进化混合算法进行了比较。为提高加工工件进行分批及分批之后子批的分配与排序效率,该问题是对不同规模的经典并行机调度问题进行求解并展示两种算法的求解,证明了灰狼差分进化混合算法在寻优性能上优于遗传差分进化混合算法,不仅具有更好的解的稳定性,而且具有更高的寻优精度。     

含忽略工序和不相关机的混合流水车间调度

研究从炼钢等生产过程提炼出的含忽略工序和不相关并行机的混合流水车间调度问题,以最小化最大完工时间为目标,建立整数规划模型,并提出结合全局搜索、自适应遗传算法和候鸟优化的遗传候鸟优化算法以求解该模型。在算法中采用与处理时间相关的全局搜索和随机程序以获得初始种群,提出自适应交叉和变异操作改进遗传算法解,在迭代进程中,引入基于工件、机器和工序位3种邻域搜索结构的候鸟优化算法更新最佳解。仿真实验中将遗传候鸟优化算法的实验结果与几种启发式算法进行对比,证明了模型和算法的有效性。     

加工时间可控的炼钢调度问题两阶段模型及优化算法

炼钢-精炼-连铸是钢铁产品的关键生产工序,其有效的调度对生产过程中减少热能消耗、提高生产效率具有重要意义.根据生产过程中工序加工时间可控性和主要工艺约束提出了分散搜索(scattcr scarch,SS)算法和数学规划相结合的两阶段求解算法.第1阶段应用SS算法基于各阶段正常的加工时间,确定炼钢-精炼生产阶段各设备的加工炉次集和各炉次的加工顺序.第2阶段将SS求得的解转化为时间约束网络图,建立了以炉次等待设备时间和设备等待炉次时间及最大完成时间最小为调度目标,工序加工时间可控的混合整数规划模型,应用CPLEX求解模型确定各炉次的加工时间和开始时间.基于国内某钢铁企业炼钢-精炼-连铸生产过程的实绩生成了14个不同规模的测试案例,对钢厂生产实绩效果与本文两阶段求解算法的优化效果进行了对比,分析了不同等待时间权重对两阶段算法性能的影响,并与采用遗传局域搜索(gcnctic local search,GLS)算法与数学规划相结合的求解算法的优化效果进行了比较.实验结果表明本文给出的模型和两阶段求解算法对加工时间可控的炼钢-精炼-连铸调度问题的优化效果很好.     

GA/LP算法求解带组换装时间的单机调度问题

以包头某钢铁线材企业实际调度问题为背景,针对带组换装时间的单机提前/拖期调度问题,建立了混合整数线性规划模型.根据模型的特点,将该问题的求解分为最优排序变量的确定和最优开工时间的确定两个层次.由于该问题是NP难的,提出了一种基于遗传算法和线性规划相结合的GA/LP算法.在算法中,利用GA在外层搜索遍历二元变量空间来优化工件排序;在内层则利用LP来确定满足约束条件的最优连续解.为了提高算法寻优性能,降低算法运行时间,提出了一种基于问题特征的种群初始化方法.最后,针对三种交叉算子在随机问题上进行了测试,实验结果表明,该算法能够有效求解问题.     

混合离散人工蜂群算法求解含不相关并行机的分布式柔性流水线调度

研究每阶段含不相关并行机的分布式柔性流水线调度问题.考虑顺序相关准备时间和工件动态到达时间,以最小化总加权提前/拖期惩罚为目标建立整数规划模型,提出一种融合离散差分进化算法、变邻域下降算法和局域搜索的混合离散人工蜂群算法以获取近优解.该算法采用基于工厂-工件号的编码以及基于机器最早空闲时间的动态解码机制,通过随机规则和均衡分派策略生成初始工厂-工件序列群,在引领蜂阶段引入离散差分进化算法产生优质工厂-工件序列,在跟随蜂阶段利用变邻域下降算法在被选择序列附近继续搜索以得到邻域序列,在侦察蜂阶段设计基于关键/非关键工厂间插入的局域搜索提高算法搜索能力.通过仿真实验测试不同规模的算例,实验结果表明,所提出的混合离散人工蜂群算法表现出较好的求解性能.     

考虑能耗与工时恶化作用下的并行机调度优化

在热处理加工环境中,工件温度随着其开工时刻的延误不断下降,为了能够正常加工不得不保温或重新加热.针对这一现象,本文考虑了能耗与工时恶化作用下的并行机调度问题,以最小化总拖期和能耗为目标构建了混合整数规划模型.由于问题的复杂性,提出了一种遗传变搜索算法,其通过遗传操作获得变邻域搜索操作的解集,而后使用变邻域结构进行寻优操作.算例测试表明：较之传统遗传算法以及数学规划器Gurobi的计算结果,所提出的算法可以有效减少综合能耗和拖期成本.     

混合离散教与学算法求解复杂并行机调度问题

针对制造行业中广泛存在的一类复杂并行机调度问题, 即带到达时间、多工序、加工约束和序相关设置时间的并行机调度问题(Parallel machine scheduling problem with arrival time, multiple operations, process restraints and sequence-dependent setup times, PMSP_AMPS), 建立问题的排序模型并提出一种混合离散教与学优化算法进行求解, 优化目标为最小化最大完工时间.首先, 根据标准教与学算法(Teaching-learning-based optimization, TLBO)中两阶段个体更新公式的特点, 在保留每一阶段个体更新公式框架不变的前提下, 对公式中具体改变实数个体或向量的每个核心操作均用所设计的排列操作进行替换, 使其可直接在离散问题解空间中执行基于标准教与学算法机理的全局搜索, 从而明显提高了原算法的全局搜索效率.其次, 采用交换操作和插入操作构造了一种简洁有效地变邻域局部搜索, 对全局搜索发现的优质解区域进行细致搜索, 从而进一步增强了算法的性能.通过对不同测试问题的仿真实验和算法比较, 验证了所提算法可有效求解PMSP_AMPS.     

混合EDA求解三阶段异构并行机装配集成调度问题

本文提出一种混合分布估计算法（hybrid estimation of distribution algorithm，HEDA）用于求解带载重约束的三阶段异构并行机集成调度问题（three-stage heterogeneous parallel machine integrated scheduling problem with capacitated constraint，THPMISP_CC），第一阶段为加工阶段，即带释放时间的多工序异构并行机调度问题；第二阶段为带载重约束的运输阶段，即多维背包优化调度问题；第三阶段为装配阶段.本文研究工件从加工、运输到装配三阶段的集成调度优化问题.首先，本文构建了THPMISP_CC的数学模型，其优化目标为三阶段整体最大完工时间（Makespan）；然后，提出的HEDA用于优化THPMISP_CC；最后，对算法运用于THPMISP_CC模型的结果进行分析和比较，验证模型的可行性及算法的有效性.     

Parallel machine scheduling in fuzzy environment with hybrid ant colony optimization including a comparison of fuzzy number ranking methods in consideration of spread of fuzziness

This paper studies parallel machine scheduling problems in consideration of real world uncertainty quantified based on fuzzy numbers. Although this study is not the first to study the subject problem, it advances this area of research in two areas: (1) Rather than arbitrarily picking a method, it chooses the most appropriate fuzzy number ranking method based on an in-depth investigation of the effect of spread of fuzziness on the performance of fuzzy ranking methods; (2) It develops the first hybrid ant colony optimization for fuzzy parallel machine scheduling. Randomly generated datasets are used to test the performance of fuzzy ranking methods as well as the proposed algorithm, i.e. hybrid ant colony optimization. The proposed hybrid ant colony optimization outperforms a hybrid particle swarm optimization published recently and two simulated annealing based algorithms modified from our previous work.     

A simulated annealing algorithm for the job shop cell scheduling problem with intercellular moves and reentrant parts

This paper addresses the problem of scheduling parts in job shop cellular manufacturing systems by considering exceptional parts that need to visit machines in different cells and reentrant parts which need to visit some machines more than once in non-consecutive manner. Initially, an integer linear programming (ILP) model is presented for the problem to minimize the makespan, which considers intercellular moves and non-consecutive multiple processing of parts on a machine. Due to the complexity of the model, a simulated annealing (SA) based solution approach is developed to solve the problem. To increase the efficiency of the search algorithm, a neighborhood structure based on the concept of blocks is applied. Subsequently, the efficiency of the ILP model and the performance of the proposed SA are assessed over a set of problem instances taken from the literature. The proposed ILP model was coded in Lingo 8.0 and the solution obtained by the proposed SA was compared to the optimal values. The computational results demonstrate that the proposed ILP model and SA algorithm are effective and efficient for this problem.     

A genetic algorithm application using fuzzy processing times in non-identical parallel machine scheduling problem

There are many scheduling problems which are NP-hard in the literature. Several heuristics and dispatching rules are proposed to solve such hard combinatorial optimization problems. Genetic algorithms (GA) have shown great advantages in solving the combinatorial optimization problems in view of its characteristic that has high efficiency and that is fit for practical application [1]. Two different scale numerical examples demonstrate the genetic algorithm proposed is efficient and fit for larger scale identical parallel machine scheduling problem for minimizing the makespan. But, even though it is a common problem in the industry, only a small number of studies deal with non-identical parallel machines. In this article, a kind of genetic algorithm based on machine code for minimizing the processing times in non-identical machine scheduling problem is presented. Also triangular fuzzy processing times are used in order to adapt the GA to non-identical parallel machine scheduling problem in the paper. Fuzzy systems are excellent tools for representing heuristic, commonsense rules. That is why we try to use fuzzy systems in this study.     

基于新型蛙跳算法的低碳柔性作业车间调度

针对低碳柔性作业车间调度问题(flexible job shop scheduling problem,FJSP),提出一种新型蛙跳算法(shuffled frog leaping algorithm,SFLA)以总碳排放最小化,该算法运用记忆保留搜索所得一定数量的最优解,并采取基于种群和记忆的种群划分方法,应用新的搜索策略如全局搜索与局部搜索的协调优化以实现模因组内的搜索,取消种群重组使算法得到简化.采用混合遗传算法和教–学优化算法作为对比算法,大量仿真对比实验验证了SFLA对于求解低碳FJSP具有较强的搜索能力和竞争力.     

Two-stage hybrid flow shop scheduling on parallel batching machines considering a job-dependent deteriorating effect and non-identical job sizes

In this paper, we investigate a specialized two-stage hybrid flow shop scheduling problem with parallel batching machines considering a job-dependent deteriorating effect and non-identical job sizes simultaneously. A novel concept of three-dimensional wasted volume based on the job normal processing time, job size, and job deteriorating rate is first proposed. Some structural properties, as well as a heuristic algorithm, are developed to solve the single parallel batching machine scheduling problem. Since the two-stage hybrid flow shop scheduling problem is NP-hard, a hybrid EDA-DE algorithm combining estimation of distribution algorithm (EDA) and differential evolution (DE) algorithm is proposed to tackle the studied problem. In addition, the Taguchi method of design of experiments (DOE) is implemented to tune the parameters of the EDA-DE. Finally, a series of computational experiments are carried out to compare the performance of the proposed hybrid EDA-DE algorithm and some recent existing algorithms from the literature, and the comparative results validate the effectiveness and efficiency of the proposed algorithm.     

基于GSPN和人工免疫算法的并行测试任务调度研究

针对ATS中并行测试任务调度复杂、难以优化的问题,提出了一种广义随机Petri网和人工免疫算法相结合的任务调度优化算法.首先对并行测试系统建立广义随机Petri网(GSPN)模型,然后将激发的变迁序列集作为并行测试任务调度路径;将免疫克隆选择算法(ICSA)应用到并行测试系统任务调度问题中,并提出一种自适应克隆选择算子,搜索最优任务调度路径,得到以测试时间最短为目标的最优任务调度方案.用某型雷达接收机并行测试系统对该算法进行仿真验证,结果表明,与改进的混合遗传算法(IHGA)相比,该算法能够便捷地得到任务调度最优序列,且测试效率更高.