基于灾变机制的预防性维护和生产调度集成优化方法

为了解决生产车间中多品种任务的生产调度与预防性维护集成优化问题,综合考虑其加工顺序、生产批量及预防性维护策略等要素,在订单充足的前提下,以总制造成本和加工时间最小化为联合优化目标,建立了生产调度与预防性维护集成优化模型。针对模型特点,在非支配排序遗传算法框架的基础上,基于灾变机制和荣誉空间,引入截断和拼接操作算子,提出一种变长度染色体单亲遗传算法对模型进行求解,并在不同参数条件和问题规模下,通过仿真实验验证了该算法解决复杂生产任务调度和预防性维护集成优化问题的有效性。     

基于柔性维护的堕化单机系统调度研究

针对单机系统,在假设生产系统为堕化系统,且生产过程中作业的加工不可中断的情况下,对考虑柔性时间窗口[[u,v]]下进行长度为[w]的周期预防性维护的调度问题进行了研究。建立了综合考虑生产调度和设备维护的混合整数规划模型,并设计了一套基于贪婪的启发式算法对所研究问题进行优化求解。通过Cplex和启发式算法求解结果的对比证明了算法可以快速、有效地解决此类问题。     

Two-machine flow shop scheduling with linear decreasing job deterioration

In this paper, we consider a two-machine flow shop scheduling problem with deteriorating jobs. By a deteriorating job, we mean that the processing time is a decreasing function of its execution start time. A proportional linear decreasing deterioration function is assumed. The objective is to find a sequence that minimizes total completion time. Optimal solutions are obtained for some special cases. For the general case, several dominance properties and some lower bounds are derived to speed up the elimination process of a branch-and-bound algorithm. A heuristic algorithm is also proposed to overcome the inefficiency of the branch-and-bound algorithm. Computational results for randomly generated problem instances are presented, which show that the heuristic algorithm effectively and efficiently in obtaining near-optimal solutions.     

面向多目标流水车间调度的多种群多目标遗传算法

针对制造型企业普遍存在的流水车间调度问题,建立了以最小化最迟完成时间和总延迟时间为目标的多目标调度模型,并提出一种基于分解方法的多种群多目标遗传算法进行求解.该算法将多目标流水车间调度问题分解为多个单目标子问题,并分阶段地将这些子问题引入到算法迭代过程进行求解.算法在每次迭代时,依据种群的分布情况选择各子问题的最好解及与其相似的个体分别为当前求解的子问题构造子种群,通过多种群的进化完成对多个子问题最优解的并行搜索.通过对标准测试算例进行仿真实验,结果表明所提出的算法在求解该问题上能够获得较好的非支配解集.     

Maintenance costs and makespan minimization for assembly permutation flow shop scheduling by considering preventive and corrective maintenance

The joint optimization of production scheduling and maintenance planning has a significant influence on production continuity and machine reliability. However, limited research considers preventive maintenance (PM) and corrective maintenance (CM) in assembly permutation flow shop scheduling. This paper addresses the bi-objective joint optimization of both PM and CM costs in assembly permutation flow shop scheduling. We also propose a new mixed integer linear programming model for the minimization of the makespan and maintenance costs. Two lemmas are inferred to relax the expected number of failures and CM cost to make the model linear. A restarted iterated Pareto greedy (RIPG) algorithm is applied to solve the problem by including a new evaluation of the solutions, based on a PM strategy. The RIPG algorithm makes use of novel bi-objective-oriented greedy and referenced local search phases to find non-dominated solutions. Three types of experiments are conducted to evaluate the proposed MILP model and the performance of the RIPG algorithm. In the first experiment, the MILP model is solved with an epsilon-constraint method, showing the effectiveness of the MILP model in small-scale instances. In the remaining two experiments, the RIPG algorithm shows its superiority for all the instances with respect to four well-known multi-objective metaheuristics.     

Flow shop问题的嵌套分区优化调度方法

介绍了嵌套分区算法(NP)的基本思想, 并用于求解流水作业优化调度问题. 算法用嵌套分区树来描述流水作业调度问题, 对可行域进行系统性分区, 然后集中搜索有优良解的区域. 在每一步迭代中, 算法跟踪最有希望的分区, 并结合启发式算法和邻域搜索来实现分区转移. 仿真实验表明, 该算法比单纯的启发式算法和邻域搜索有较好的寻优能力.     

多目标军用飞机维修作业调度优化研究

为提升维修作业与现代战机的适应程度,对军用飞机维修作业调度模型构建与调度优化算法设计进行探讨。在沿用柔性作业车间调度问题的形式化描述构建维修作业调度模型的基础上,选取遗传算法对执行步骤进行设计,引入耦合算子重新调整工序排序部分染色体以避免染色体违背耦合约束无法解码的情况发生,并采用维修作业调度案例与Brandimarte测试数据验证多目标调度优化算法的适用性与优化性。维修作业调度模型构建与调度优化算法的探讨促进维修管理的精细化,为调度相关领域的深入研究拓宽思路。     

Two-machine flow shop problems with a single server

Scheduling models that allow the handling of pre-operational setup have been a source of major interests because of their practical relevance and theoretical impacts. Two-stage flow-lines have drawn much attention to researchers as they are simple, yet practical and can be easily extended to represent more complex situations. In this paper, two-machine flow-shop problems with a single setup server are surveyed. These problems have been shown to be NP-complete with special cases that are polynomial-time solvable. Several heuristics are proposed to solve the problems in general case, including simulated annealing, Tabu search, genetic algorithms, GRASP, and other hybrids. The results on small inputs are compared with the optimal solutions and results on large inputs are compared to a lower bound. Experiments show that the heuristics developed, obtain nearly optimal solutions.     

Two-machine open shop scheduling with secondary criteria

This paper considers two-machine open shop problems with secondary criteria where the primary criterion is the minimization of makespan and the secondary criterion is the minimization of the total flow time, total weighted flow time, or total weighted tardiness time. In view of the strongly NP-hard nature of these problems, two polynomially solvable special cases are given and constructive heuristic algorithms based on insertion techniques are developed. A strongly connected neighborhood structure is derived and used to develop effective iterative heuristic algorithms by incorporating iterative improvement, simulated annealing and multi-start procedures. The proposed insertion and iterative heuristic algorithms are empirically evaluated by solving problem instances with up to 80 jobs.     

流水车间调度问题的快速多目标混合进化算法

针对最大完工时间最小和总流经时间最小的双目标流水车间调度问题,提出一种快速多目标混合进化算法。算法将矢量评价遗传算法的采样策略与一种新的基于Pareto支配与被支配关系的适应度函数的采样策略进行了融合。新的采样策略弥补了矢量评价遗传算法(VEGA)采样策略的不足。VEGA善于搜索Pareto前沿面的边缘区域,但却忽略了Pareto前沿面的中心区域,而新的采样策略则倾向于Pareto前沿面的中心区域。这两种机制的融合保证了混合算法能够快速平稳地向Pareto前沿区域收敛。此外,由于混合采样策略不需要考虑距离,使得算法效率也得到了很大的提升。在对Taillard基准测试集进行的仿真实验结果显示,相对于非支配排序遗传算法(NSGA-Ⅱ)和强度Pareto进化算法(SPEA2),该快速多目标混合进化算法在收敛性和分布性两方面都有所提高,并且算法的效率也得到了改进。所提出的混合算法能够更好地解决双目标的流水车间调度问题。     

Two-machine flow shop scheduling problem with blocking,multi-task flexibility of the first machine,and preemption

Blocking flow shop scheduling problem has been extensively studied in recent years; however, some applications mentioned for this problem have some additional characteristics that have not been well considered. Multi-task flexibility of machines and preemption are two of such characteristics. Multi-task flexible machines are capable of processing the operations of at least one other machine in the system. In addition, if preemption is allowed, the solution space grows, and solutions that are more efficient may be obtained. In this study, the two-machine flow shop scheduling problem with blocking, multi-task flexibility of the first machine, and preemption is investigated by considering the minimization of makespan as criterion. It is proved that the complexity of the problem is strongly NP-hard. Because of preemption and multi-task flexibility, there are infinite schedules for each sequence; however, it is shown that a dominant schedule can be defined for each sequence. Two mathematical models are proposed for optimally solving the small-sized instances. Furthermore, a variable neighborhood search algorithm (VNS) and a new variant of it, namely, dynamic VNS (DVNS), are presented to find high quality solutions for large-sized instances. Unlike the VNS algorithm, the DVNS algorithm does not need tuning for the shaking phase. Nevertheless, computational results show that DVNS has even a slightly better performance. The VNS and DVNS algorithms are also compared with some of the best-performing metaheuristics already developed for the flow shop scheduling problem with blocking and minimization of makespan as criterion. Computational results reveal that both algorithms are superior to the others for large-sized instances.     

A genetic algorithm for permutation flow shop scheduling under make to stock production system

The permutation flow shop scheduling is a well-known combinatorial optimization problem that arises in many manufacturing systems. Over the last few decades, permutation flow shop problems have widely been studied and solved as a static problem. However, in many practical systems, permutation flow shop problems are not really static, but rather dynamic, where the challenge is to schedule n different products that must be produced on a permutation shop floor in a cyclical pattern. In this paper, we have considered a make-to-stock production system, where three related issues must be considered: the length of a production cycle, the batch size of each product, and the order of the products in each cycle. To deal with these tasks, we have proposed a genetic algorithm based lot scheduling approach with an objective of minimizing the sum of the setup and holding costs. The proposed algorithm has been tested using scenarios from a real-world sanitaryware production system, and the experimental results illustrates that the proposed algorithm can obtain better results in comparison to traditional reactive approaches.     

Joint optimization of job scheduling and maintenance planning for a two-machine flow shop considering job-dependent operating condition

Two-machine flow shops are widely adopted in manufacturing systems. To minimize the makespan of a sequence of jobs, joint optimization of job scheduling and preventive maintenance (PM) planning has been extensively studied for such systems. In practice, the operating condition (OC) of the two machines usually varies from one job to another because of different processing covariates, which directly affects the machines’ failure rates, PM plans, and expected job completion times. This fact is common in many real systems, but it is often overlooked in the related literature. In this study, we propose a joint decision-making strategy for a two-machine flow shop with resumable jobs. The objective is to minimize the expected makespan by taking into account job-dependent OC. We consider two situations. In the first situation, where the failure rate of a machine under a fixed OC is constant, a hybrid processing time model is proposed to obtain the optimal job sequence based on the Johnson's law. For the second situation, where the failure rate of a machine is time-varying, the job sequence and PM plan are jointly optimized. An enumeration method is adopted to find the optimal job sequence and PM plan for a small-scale problem, and a genetic algorithm-based method is proposed to solve a large-scale problem. Numerical examples are provided to demonstrate the necessity of considering the effect of job-dependent OC and the effectiveness of the proposed method in handing such joint decision-making problems in manufacturing systems.     

求解流水线调度问题的万有引力搜索算法

研究了以最大完工时间为目标的流水线调度问题,使用万有引力算法求解调度问题,提出了一种最大排序规则,利用物体间各个位置分量值存在的大小次序关系,并结合随机键编码的方法产生,将物体的连续位置转变成了一个可行的调度方案;提出了一种边界变异的策略使得越界的物体不再聚集在边界上,而是分布在边界附近的可行空间内,从而增加种群的多样性;结合交换算子和插入算子提出了一种新的局部搜索算法,有效地避免了算法陷入局部最优值,进一步提高了解的质量.最后证明了算法的收敛性,并且计算了算法的时间复杂度和空间复杂度,仿真实验说明了所得算法的有效性.     

并行流程车间调度问题及其概率学习进化算法

并行Flowshop调度问题兼有并行机器和流程车间调度问题的特点,是一类新型的调度问题.针对最小化最大完工时间目标函数,建立了一般并行Flowshop调度问题的整数规划模型.鉴于问题的求解复杂性,设计了基于概率学习的求解算法.对随机生成的测试问题进行求解,实验结果显示出该算法求解并行Flowshop调度问题的良好潜能.     

最优子种群遗传算法求解柔性流水车间调度问题

为了验证最优子种群遗传算法在解决柔性流水车间调度问题时相比于传统遗传算法的优越性,分析了柔性流水车间调度问题的特点,并运用一种新的编码方法和新的遗传算法求解了该问题。考虑到最优个体保护策略法对复杂问题容易使种群收敛陷入局部最优解,为了提高精度、加快较优个体的产生并避免陷入局部最优解,首先提出了一种合理、全面的编码方法,并运用最优子种群遗传算法来求解柔性流水车间调度问题。最后运用实例验证了最优子种群遗传算法的有效性、优越性和编码方式的合理性。     

一种求解无等待流水车间调度优化的启发式算法

提出了一种新的启发式算法,用于求解无等待流水车间调度问题的总流水时间指标。该算法命名为标准差启发,基于著名的NEH启发算法。首先阐述了总流水时间指标;其次描述了标准差启发算法的过程;最后用标准差启发算法求解标准实验案例,通过实验并与其他启发式算法比较,验证了标准差启发算法在求解无等待流水车间调度问题总流水时间指标的有效性。     

Minimizing total completion time for re-entrant flow shop scheduling problems

In this paper, we study re-entrant flow shop scheduling problems with the objective of minimizing total completion time. In a re-entrant scheduling problem, jobs may visit some machines more than once for processing. The problem is NP-hard even for machine number m=2. A heuristic algorithm is presented to solve the problem, in which an effective k-insertion technique is introduced as the improvement strategy in iterations. Computational experiments and analyses are performed to give guidelines of choosing parameters in the algorithm. We also provide a lower bound for the total completion time of the optimal solution when there are only two machines. Objective function values of the heuristic solutions are compared with the lower bounds to evaluate the efficiency of the algorithm. For randomly generated instances, the results show that the given heuristic algorithm generates solutions with total completion times within 1.2 times of the lower bounds in most of the cases.     

Modelling and scheduling multi-objective flow shop problems with interfering jobs

This paper studies multi-objective flow shop scheduling problems with interfering jobs. That is, there are two sets of jobs and each of which has its own objective. Some jobs are scheduled so as to minimize makespan while the others are to minimize total tardiness. In this case, the problem was mathematically modeled by a mixed integer linear program. Then, a novel biogeography-based optimization was developed to solve the problem. To evaluate the algorithm, its performance was compared with three well-known algorithms in the literature. The results of the present study show that the proposed algorithm outperforms the other tested algorithms.