柔性工作车间调度问题的多目标优化方法研究

针对各工件日标不同的多目标柔性作业车间调度问题,构建了以加工成本、加工质量及制造工期为目标函数的柔性作业车间调度多日标优化数学模型.针对传统的加权系数遗传算法不能很好地解决柔性作业车间调度多目标优化问题,提出采用改进的强度Pareto进化算法,对柔性作业车间调度问题进行多目标优化,从而得出柔性车间调度问题的Pareto综合最优解.最后,结合项目实施,以某大型空分装备企业的车间调度为例,证明了文中提出的方法能很好地解决柔性工作车间调度的多目标优化问题.     

并行机作业车间等量分批多目标优化调度

为降低并行机作业车间等量分批多目标优化调度问题的复杂度,提高优化效率,提出了一种基于仿真技术和改进非支配排序遗传算法的分步优化方法.建立了一类以完工时间最短和总制造成本最低为优化目标的并行机作业车间等量分批多目标优化调度模型;将各产品进行等量分批,以Witness为仿真平台建立并行机作业车间等量分批生产仿真模型,通过组合仿真优化得到产品理想的等量分批方案,从而将原问题转化为并行机作业车间多目标优化调度问题;设计了一种改进的非支配排序遗传算法,对并行机作业车间多目标优化调度进行求解.通过算例分析验证了该方法的有效性.     

订单型企业多目标生产计划的制定及其调度

针对制定订单式小批量生产计划问题,提出了一种使用动态随机投入产出函数来制定多目标生产计划的方法。针对生产调度问题,提出了联合使用最长加工时间优先(LPT)与遗传算法(GA)的混合遗传算法(HGA)来求解混合流水线的调度,并给出了一种新的编码方法,选择了相应的交叉和变异方法。研究结果表明,该计划制造方法能较好地满足订单型企业的随机性要求,而且生产计划编制效率高。该编码方法在保证染色体合法性的同时也保证了算法本身的随机性。某轧辊厂的实际案例分析结果也验证了所提出的订单型企业多目标生产计划的制定及其调度方法的可行性。     

药房自动化设备出药装置的结构优化设计

针对药房自动化设备容易出现的不出药问题,提出一种药房自动化设备出药装置的工作原理和结构形式。利用MATLAB软件对其中关键的翻板结构进行了优化仿真,得出合理的设计参数。最后,通过试验验证优化后的结构能够实现药品的快速可靠发放,达到系统预期指标。     

资源约束下产品开发任务调度的多目标优化

鉴于产品开发任务调度过程中存在资源约束问题和学习与遗忘效应,需要对多个目标进行优化决策,通过定义资源平均利用率并提出学习遗忘效应矩阵,结合耦合设计的多阶段迭代模型,以各阶段资源利用率为约束条件,建立资源约束下考虑学习与遗忘效应的任务调度时间与成本的多目标优化数学模型。采用带精英策略的非支配排序遗传算法求解得出Pareto最优解集,并采用改进的多目标理想点法对该解集进行选优,得到最优任务调度方案。以某电动汽车的开发过程为例,验证了该优化模型能够减小产品开发时间,降低产品开发成本,提高总资源利用率。     

Multi-objective simulation optimization for selection and determination of order quantity in supplier selection problem under uncertainty and quality criteria

Simulation optimization is providing solutions to practical stochastic problems. Supplier selection is one of the most important decisions that determine the survival of an organization. In this paper, a novel multi-objective simulation optimization method to make decisions on selecting the suppliers and determining the order quantities is proposed. Regarding the fact that a real supply chain is multi-objective with uncertain parameters and includes both quantitative and qualitative variables, the proposed method considers these points and is applicable to real-world problems. This method also considers supplier selection and order quantity allocation to each supplier, which are totally related, as an integrated model. The proposed method consists of four basic modules: Cuckoo Optimization Algorithm (COA), Discrete Event Simulation (DES), Supply Chain Model (SCM), and Generalized Data Envelopment Analysis (GDEA). Unlike many multi-objective methods, the proposed method is not limited to the number of objective functions and this is one of its main benefits. It also pays attention to the efficiency of the organization and, at the same time, finding inputs which result in best output amounts. This method, in addition to the convergence criterion, pays special attention to the dispersion of the Pareto frontier as the second criterion for choosing the good solutions. For implementation of the proposed method, the numerical results for the problem of supplier selection in multi-product, multi-customer modes, and uncertain and qualitative variables are discussed and the Pareto frontiers are presented. The proposed method in this paper is compared with a similar method, and the results show the efficiency of the proposed method.     

无成组技术条件下流水车间调度的多目标优化

针对有工件组调整时间的流水车间调度问题,提出了无成组技术假设条件下的多目标优化模型,并设计了一种进化计算与局部搜索结合的混合遗传算法.模型的目标函数是最小化最大完工时间和最大拖期.在局部搜索过程中,根据问题的特征定义了两种邻域结构,采取两阶段搜索策略,以提高算法的优化搜索效率.进化过程中,采用基于个体的累计排序数和密度值的适应度分配方法,以保持群体多样性,并采取精英保留策略,以保证解的收敛性.通过测试问题和实际问题的实验以及与其他算法的比较,验证了所提模型和算法的有效性.     

考虑资源约束的预制构件多目标生产调度优化

为优化装配式建筑预制构件生产调度问题,从生产供应的角度对预制构件生产流程进行分析,同时考虑生产过程中的资源约束,构建了以生产完工时间和惩罚成本为目标的预制构件生产调度数学模型.设计了一种新颖的多目标混合共生生物搜索算法对模型进行求解,以合理安排预制构件的生产顺序和资源配置,达到降低成本、提高生产效率的目的 .通过装配式住宅项目的 一个实例验证了模型和算法的有效性.     

基于改进MOEA/D的多目标置换流水车间调度问题

针对工期和总流水时间的两目标置换流水车间调度问题,提出一种改进的基于分解的多目标进化算法(MOEA/D).为了改进非支配解集的质量,提高算法效率,在MOEA/D中嵌入分组和统计学习机制提出一种两阶段局部搜索策略改进外部存档.利用基于距离的替换策略更新种群,提高种群的多样性,保证了分组机制的有效性.基于Taillard标准测试问题的实验结果表明,所提出的改进MOEA/D算法明显优于传统MOEA/D、NS-GA-Ⅱ、MEDA/D-MK等算法.     

求解Job shop调度的协同优化算法

针对量子粒子群算法、遗传算法在求解车间调度存在的局部收敛的问题,提出用量子粒子群算法与遗传算法相结合的协同优化方法求解该问题。该算法采用量子粒子群算法与遗传算法的并行搜索结构,通过迁移算子把各个种群联系起来。仿真结果表明,该算法收敛速度快,且具有较高的求解质量。     

Ant colony optimization technique for the sequence-dependent flowshop scheduling problem

In the real world, production scheduling systems, usually optimal job scheduling, requires an explicit consideration of sequence-dependent setup times. One of the most important scheduling criteria in practical systems is makespan. In this paper, the author presents an ant colony optimization (ACO) algorithm for the sequence-dependent permutation flowshop scheduling problem. The proposed ACO algorithm benefits from a new approach for computing the initial pheromone values and a local search. The proposed algorithm is tested on randomly generated problem instances and results indicate that it is very competitive with the existing best metaheuristics.     

Bi-objective optimization of maintenance scheduling for power systems

Enhance the quality of energy production in power generating stations and reducing its cost have become of paramount importance. One of the methods to reach that goal is by minimizing the maintenance scheduling time. For this purpose, a new competitive mechanism, based on a modified genetic algorithm (MGA), has been proposed to perform the preventive maintenance (PM) scheduling. Firstly, a mono-objective optimization (makespan) has implemented, and the results were quite good. Secondly, and in order to benefit from the waste time, a bi-objective optimization was developed to find a trade-off between makespan and training time of operators. Finally, the MGA-based maintenance scheduling was tested on a hybrid renewable power system (HRPS), that uses photovoltaic modules and a fuel cell (PV/FC) as sources and the telecommunication platform as load, the obtained results have proved the high efficiency of the proposed MGA-based maintenance scheduling.     

求解作业车间调度问题的广义粒子群优化算法

为克服传统粒子群优化算法在解决组合优化问题上的局限性,分析了其优化机理,并在此基础上提出了广义粒子群优化模型。按照此模型提出了一种求解作业车间调度问题的广义粒子群优化算法。在本算法中,利用遗传算法中的交叉操作作为粒子间的信息交换策略,利用遗传算法中的变异操作作为粒子的随机搜索策略,而粒子的局部搜索策略则采用禁忌搜索来实现。为了控制粒子的局部搜索以及向全局最优解的收敛,迭代过程中交叉概率以及禁忌搜索的最大步长都是动态变化的。实验结果表明,本算法可有效地求解作业车间调度问题,验证了广义粒子群优化模型的合理性。     

基于PSO的车间柔性调度计算

在不确定车间信息环境下的调度策略问题中,提出多目标车间柔性调度的规划模型,并应用粒子群算法进行求解,给出具体的算法应用过程.实验结果验证该模型和算法的有效性及实用性.     

A hybrid particle swarm optimization approach for the job-shop scheduling problem

Managing multiple projects is a complex task. It involves the integration of varieties of resources and schedules. The researchers have proposed many tools and techniques for single project scheduling. Mathematical programming and heuristics are limited in application. In recent years non-traditional techniques are attempted for scheduling. This paper proposes the use of a heuristic and a genetic algorithm for scheduling a multi-project environment with an objective to minimize the makespan of the projects. The proposed method is validated with numerical examples and is found competent.     

An improved particle swarm optimization for the resource-constrained project scheduling problem

In this paper, an improved particle swarm optimization (PSO) algorithm is proposed for the resource-constrained project scheduling problem (RCPSP) which is widely applied in advanced manufacturing, production planning, and project management. The algorithm treats the solutions of RCPSP as particle swarms and employs a double justification skill and a move operator for the particles, in association with rank-priority-based representation, greedy random search, and serial scheduling scheme, to execute the intelligent updating process of the swarms to search for better solutions. The integration combines and overhauls the characteristics of both PSO and RCPSP, resulting in enhanced performance. The computational experiments are subsequently conducted to set the adequate parameters and compare the proposed algorithm with other approaches. The results suggest that the proposed PSO algorithm augments the performance by 9.26, 16.17, and 10.45 % for the J30, J60, and J120 instances against the best lower bound-based PSO currently available, respectively. Moreover, the proposed algorithms demonstrate obvious advantage over other proposals in exploring solutions for large-scale RCPSP problems such as the J60 and J120 instances.     

采用遗传算法的混流装配线平衡多目标优化

针对混流装配线操作人数已知、生产节拍待定情况下的平衡问题,研究了混流装配线第二类平衡问题的优化方法,以均衡各工作站的加权平均负荷、最小化生产节拍为优化目标,建立了求解加权平均负荷与生产节拍加权和的数学模型,并运用遗传算法对其进行求解.各优化目标数量级的统一,简化了算法求解过程,在保证解的有效性的同时提高了算法求解效率.算例分析验证了算法的有效性.     

Heuristics for assembly operation scheduling problem in flexible assembly systems

In this paper we address the problem of scheduling n assembly operations with in-tree constraints on m unrelated parallel workstations in flexible assembly systems, which we call the assembly operation scheduling problem (AOSP). No preemption of assembly operations is allowed and the primary objective is to minimize the maximum completion time. This problem is equivalent to R|tree|C_max. For this notorious NP-hard problem, four heuristic algorithms including decomposition (DECOMP), earliest completion time (ECT), shortest processing time (SPT), and earliest starting time(EST) heuristic are proposed and their performances are comparatively investigated. DECOMP uses a decomposition technique to practically solve this problem. The assembly operation tree is decomposed and then AOSP is reduced to a set of subproblems of R||C_max and R,r_j|r_i|C_max. Two efficient heuristics were proposed for the reduced subproblems. The other three heuristics basically use the machine selection rules to determine the machine for processing the current operation. Of these heuristics, DECOMP showed the best performance in terms of quality of schedule. Computational results show that all the proposed algorithms except the EST heuristic perform quite well in terms of both quality of solution and computation time.     

Design automation and optimization of assembly sequences for complex mechanical systems

Assembly sequence design for a mechanical system can have significant impact on manufacturing cost and product quality. Traditionally, such a design process is largely based on experience and best practices, often ineffective and non-optimal as the system becomes complex. This paper proposes a new, systematic method for automatic assembly sequence design and optimization. Key elements include assembly modeling, sequence planning, locating scheme optimization, dimensional quality evaluation, and optimization. First, a directed graph and an assembly matrix are employed to describe the assembly relation of a system. Then, the feasible assembly sequences are generated through layered subassembly detection of adjacency matrices, filtered by engineering constraints. To evaluate quickly the assembly quality and compare the influences of different locating point schemes, a linear 3D variation propagation analysis model with deterministic locating principle is introduced. The optimal locating scheme is then selected using a genetic algorithm with the least variation propagation. Finally, the assembly dimensional quality for different sequences is evaluated and the optimal assembly sequences are achieved through genetic algorithms. A case of automotive body side assembly is presented to illustrate the whole methodology.