A GA-based approach to machining operation sequencing for prismatic parts

This paper presents a study of using a genetic algorithm (GA) method to select the machining operation sequence for prismatic parts. Four types of process planning rules including precedence rules, clustering rules, adjacent order rules and optimization rules are considered and are encompassed quantitatively in the fitness calculations for alternative operation sequences. The impact of variations of the rules on the result of operation sequencing and that of GA parameters on the solution efficiency are discussed through analysis of examples and experiments. The proposed genetic algorithm proves effective for machining operation sequencing of prismatic parts, by incorporating various production environment considerations into process planning.     

A CAPP framework with optimized process parameters for rotational components

Process planning, as a critical stage integrating the design and manufacturing phase in a manufacturing environment, has been automated to meet the needs for higher productivity and lower production cost. Being an input to various systems such as scheduling and routing, process planning results are of great importance in the manufacturing stage. Though feature extraction and sequence optimization have been given much attention, the process parameters are rarely dealt with. This paper focuses on the development of a new generative computer aided process planning (CAPP) framework for rotational components. The developed framework includes modules for feature extraction based on CAD application programming interfaces, determination of the optimum sequence and generation of optimum process parameters. The optimization of the machining operations is achieved using the evolutionary technique. The approach resulted in the reduction and prediction of machining time and cost. The framework is demonstrated with a case study.     

Convergence Properties of Genetic Algorithmsin a Wide Variety of Noisy Environments

Random noise perturbs objective functions in practical optimization problems, and genetic algorithms (GAs) have been proposed as an effective optimization tool for dealing with noisy objective functions. In this paper, we investigate GAs in a variety of noisy environments where fitness perturbation can occur in any form-for example, fitness evaluations can be concurrently disturbed by additive and multiplicative noise. We reveal the convergence properties of GAs by constructing and analyzing a Markov chain that explicitly models the evolution of the algorithms in noisy environments. We compute the one-step transition probabilities of the Markov chain and show that the chain has only one positive recurrent communication class, which is also aperiodic. Based on this property, we establish a condition that is both necessary and sufficient for GAs to eventually (i.e., as the number of iterations goes to infinity) find a globally optimal solution with probability 1. We also identify a condition that is both necessary and sufficient for GAs to eventually with probability 1 fail to find any globally optimal solution. Furthermore, in all the noisy environments, our analysis shows that the chain has a stationary distribution that is also its steady-state distribution. Based on this property and the transition probabilities of the chain, we examine the number of iterations sufficient to ensure with at least a specified probability that GAs select a globally optimal solution upon termination.     

基于模糊神经网络的公差-成本模型研究

研究了面向制造环境的公差-成本模型。对制造环境中有关的加工因素进行分类及模糊化处理,构造了加工因素的成本模糊影响系数,并以此系数和零件公差作为输入,建立了基于模糊神经网络的公差-成本模型。此模型在表征加工成本、公差关系上的精度较高,更适应面向制造公差设计的要求。     

Availability consideration in the optimal selection of multiple-aspect RMS configurations

Machine availability has a profound influence on the performance of manufacturing systems. This paper extends a model for optimizing reconfigurable manufacturing systems (RMS) configurations with multiple-aspects to incorporate the effect of machine availability using the universal generating function (UGF). Two powerful meta-heuristic optimization techniques, namely genetic algorithms (GAs) and tabu search (TS), are used for optimizing the capital cost and system availability of the RMS configurations. The optimized configurations can handle multiple-parts and their structure is that of flow lines allowing paralleling of identical machines in each production stage. The various aspects considered in the RMS configurations include arrangement of machines, equipment selection and assignment of operations. A case study is presented and implementation of the optimization model is carried out using MATLAB software. The results of using both GAs and TS to solve the problem are then reported and compared for validation. Analysis of different cases of availability consideration including infinite and no buffer capacity is performed and results are compared to those obtained when machine availability is not considered. It has been shown that considering availability affects the optimal configuration selection and increases the required equipment. This increases the costs of the near-optimal configurations obtained especially in the case without buffers. The presented model can support the manufacturing systems configuration selection decisions at both the initial design and reconfiguration stages.     

遗传算法在网络计划费用优化中的应用

人们对网络计划费用优化问题的研究主要是基于传统的网络计划技术和数学规划的方法，但是这些方法都存在一些缺陷。为解决传统优化方法中存在的问题，将经过改进的遗传算法应用于费用优化当中，从而可以得到一种更精确、求解效率更高的优化方法。     

A parallel genetic algorithm for dynamic cell formation in cellular manufacturing systems

Instead of using expensive multiprocessor supercomputers, parallel computing can be implemented on a cluster of inexpensive personal computers. Commercial accesses to high performance parallel computing are also available on the pay-per-use basis. However, literature on the use of parallel computing in production research is limited. In this paper, we present a dynamic cell formation problem in manufacturing systems solved by a parallel genetic algorithm approach. This method improves our previous work on the use of sequential genetic algorithm (GA). Six parallel GAs for the dynamic cell formation problem were developed and tested. The parallel GAs are all based on the island model using migration of individuals but are different in their connection topologies. The performance of the parallel GA approach was evaluated against a sequential GA as well as the off-shelf optimization software. The results are very encouraging. The considered dynamic manufacturing cell formation problem incorporates several design factors. They include dynamic cell configuration, alternative routings, sequence of operations, multiple units of identical machines, machine capacity, workload balancing, production cost and other practical constraints.     

Combining neural networks and genetic algorithms for optimising the parameter design of the inter-metal dielectric process

An inter-metal dielectric (IMD) is deposited between metal layers to provide isolation capability to a device and separate the different metal layers that are unnecessary in the conduction of electricity. Owing to the complicated input/response relationships of the IMD process, the void problem results in electric leakage and causes wafer scraping. In the current study, we combined neural networks, genetic algorithms (GAs) and the desirability function in order to optimise the parameter settings of the IMD process. Initially, a backpropagation (BP) neural network was developed to map the complex non-linear relationship between the process parameters and the corresponding responses. Moreover, the desirability function and GAs were employed to obtain the optimum operating parameters in respect to each response. The implementation of the proposed approach was carried out in a semiconductor manufacturing company in Taiwan, and the results illustrate the practicability of the proposed approach.     

模块化生产在复杂系统制造过程的运用研究

基于复杂系统制造过程对完善生产资料准备、缩短制造周期、增强生产计划控制力、提升高端制造业发展水平的迫切要求,根据模块化理论、制造工艺、项目管理、最优化理论等方法,构建复杂系统制造模块顺序网络图,并建立求解制造周期的整数规划模型,最后举出实例,运用优化求解软件CPLEX得出整个系统制造周期及对应的关键制造路径,运用求解结果制定生产资料准备计划及生产计划,发现模块化技术实行后,生产计划及生产资料准备计划实现最大并行化开展,且生产资料准备先于生产计划开展前完成,整个系统制造周期大大缩短,验证了模型的有效性、准确性和优越性。     

A Genetic Algorithm for Target Tracking in FLIR Video Sequences Using Intensity Variation Function

Automatic target tracking in forward-looking infrared (FLIR) imagery is a challenging research area in computer vision. This task could be even more critical when real-time requirements have to be taken into account. In this context, techniques exploiting the target intensity profile generated by an intensity variation function (IVF) proved to be capable of providing significant results. However, one of their main limitations is represented by the associated computational cost. In this paper, an alternative approach based on genetic algorithms (GAs) is proposed. GAs are search methods based on evolutionary computations, which exploit operators inspired by genetic variation and natural selection rules. They have been proven to be theoretically and empirically robust in complex space searches by their founder, J. H. Holland. Contrary to most optimization techniques, whose goal is to improve performances toward the optimum, GAs aim at finding near-optimal solutions by performing parallel searches in the solution space. In this paper, an optimized target search strategy relying on GAs and exploiting an evolutionary approach for the computation of the IVF is presented. The proposed methodology was validated on several data sets, and it was compared against the original IVF implementation by Bal and Alam. Experimental results showed that the proposed approach is capable of significantly improving performances by dramatically reducing algorithm processing time.     

A genetic algorithm with memory for optimal design of laminated sandwich composite panels

This paper is concerned with augmenting genetic algorithms (GAs) to include memory for continuous variables, and applying this to stacking sequence design of laminated sandwich composite panels that involves both discrete variables and a continuous design variable. The term “memory” implies preserving data from previously analyzed designs. A balanced binary tree with nodes corresponding to discrete designs renders efficient access to the memory. For those discrete designs that occur frequently, an evolving database of continuous variable values is used to construct a spline approximation to the fitness as a function of the single continuous variable. The approximation is then used to decide when to retrieve the fitness function value from the spline and when to do an exact analysis to add a new data point for the spline. With the spline approximation in place, it is also possible to use the best solution of the approximation as a local improvement during the optimization process. The demonstration problem chosen is the stacking sequence optimization of a sandwich plate with composite face sheets for weight minimization subject to strength and buckling constraints. Comparisons are made between the cases with and without the binary tree and spline interpolation added to a standard GA. Reduced computational cost and increased performance index of a GA with these changes are demonstrated.     

基于粒子群算法优化 BP 神经网络的色彩空间转换

目的研究基于粒子群算法优化BP神经网络对显示器色彩空间转换的预测准确性的方法。方法主要通过数据归一化处理、改进最大限制速度、惯性常数和适应度函数来优化BP神经网络的权值和阈值,以缩小其分布范围,再用BP神经网络法进行色差预测。结果改进粒子群算法优化BP神经网络预测模型,测试20次得到色块平均色差为2.8526,最小平均色差为2.0453。结论该方法大大降低了BP神经网络预测模型陷入局部极小值的可能性,对显示器色彩空间转换具有较好的非线性拟合能力和更高的预测准确性。     

A simulated annealing-based optimization algorithm for process planning

Computer-aided process planning (CAPP) in the past typically employed knowledge-based approaches, which are only capable of generating a feasible plan for a given part based on invariable machining resources. In the field of concurrent engineering, there is a great need for process planning optimization. This paper describes an approach that models the constraints of process planning problems in a concurrent manner. It is able to generate the entire solution space by considering multiple planning tasks, i.e. operations (machine, tool and tool approach direction), selection and operations sequencing simultaneously. Precedence relationships among all the operations required for a given part are used as the constraints for the solution space. The relationship between an actual sequence and the feasibility of applying an operation is also considered. An algorithm based on simulated annealing (SA) has been developed to search for the optimal solution. Several cost factors including machine cost, tool cost, machine change cost, tool change cost and set-up change cost can be used flexibly as the objective function. The case study shows that the algorithm can generate highly satisfying results.     

PSO优化多尺度一维卷积神经网络的风机基础螺栓松动诊断

为在非经验指导下获取多尺度一维卷积神经网络中卷积核数目和尺度最优参数,实现风机基础螺栓松动智能诊断,提出粒子群优化(particle swarm optimization, PSO)多尺度一维卷积神经网络的风机基础螺栓松动诊断方法。首先,获取风机一维原始振动信号,划分训练集与验证集;然后,将多尺度一维卷积神经网络中卷积核数目和尺度作为PSO的粒子,以验证精度作为适应度值,根据适应度值更新粒子速度和位置,经训练后获得最优卷积核数目和尺度参数下的多尺度一维卷积神经网络;最后,输入测试样本,得到风机基础螺栓松动诊断结果。在稳定转速和升降速下进行风机基础螺栓松动诊断试验,结果表明,PSO优化多尺度一维卷积神经网络的风机基础螺栓松动诊断方法可在非经验指导下获取最优参数,可从一维原始信号中提取出有效松动特征,具备良好的松动诊断效果。     

重载并联六维力传感器及静态标定

针对传感器重载小尺寸需求,提出一种具有混合分支的重载并联六维力传感器,分析了其结构特点和测量原理。搭建了重载并联六维力传感器标定系统,为改善维间耦合及制造误差等对测量精度产生的影响,从标定算法及模型优化方面对其进行了研究。分别利用最小二乘法和BP神经网络算法对加载实验数据进行了处理,分析结果表明BP神经网络算法要明显优于最小二乘法,并通过数据随机分组测试验证了结果的正确性。基于BP神经网络,提出了一种基于人工鱼群算法的BP神经网络算法,并采用优化后的BP神经网络标定算法对实验数据进行了计算分析,结果表明优化后的BP神经网络计算结果较好且稳定,不易陷入局部极值。     

PSO优化多尺度一维卷积神经网络的风机基础螺栓松动诊断

为在非经验指导下获取多尺度一维卷积神经网络中卷积核数目和尺度最优参数,实现风机基础螺栓松动智能诊断,提出粒子群优化(particle swarm optimization, PSO)多尺度一维卷积神经网络的风机基础螺栓松动诊断方法。首先,获取风机一维原始振动信号,划分训练集与验证集;然后,将多尺度一维卷积神经网络中卷积核数目和尺度作为PSO的粒子,以验证精度作为适应度值,根据适应度值更新粒子速度和位置,经训练后获得最优卷积核数目和尺度参数下的多尺度一维卷积神经网络;最后,输入测试样本,得到风机基础螺栓松动诊断结果。在稳定转速和升降速下进行风机基础螺栓松动诊断试验,结果表明,PSO优化多尺度一维卷积神经网络的风机基础螺栓松动诊断方法可在非经验指导下获取最优参数,可从一维原始信号中提取出有效松动特征,具备良好的松动诊断效果。     

Defining specialized design similarity measures

Similar product designs resemble each other and have similar attributes and characteristics. When examining a product design to create its process plan, design its fixture, or estimate its manufacturing cost, manufacturing engineers often identify one or more similar products that the factory has manufactured in the past. They may do this from memory, from a file cabinet, or from a database (in a product data management system). Then, they retrieve the process plan (or fixture or cost estimate) for the similar product and modify it for the new product. Manufacturing and process planning experts use a complex set of rules, guidelines, and other knowledge to determine how similar two products are. Computer-aided process planning tools, however, generally use simpler, less sophisticated procedures for determining similarity. These traditional procedures may be inappropriate for specific settings. This paper presents an approach for developing function-specific design similarity measures. Such a measure explicitly exploits the specific need for similar products and thus improves variant approaches for process planning, fixture planning, and manufacturability evaluation. The approach is applied to a specific fixture planning domain.     

Cost-tolerance analysis model based on a neural networks method

The purpose of tolerance design in product components is to produce a product with the least manufacturing cost possible, while meeting all functional requirements of the product. The product designer and process planner must fully understand the process accuracy and manufacturing cost of all kinds of manufacturing process to perform a good process plan job. Usually, the cost-tolerance model is constructed by a linear or non-linear regression analysis based on the data of the cost-tolerance experiment and to derive the correlation curve between the two. Though these correlation curves can show the relationship between manufacturing cost and tolerance, a fitting error is inevitable. In particular, there is considerable discrepancy in terms of the non-experimental data. A cost-tolerance analysis model based on a neural networks method is proposed. The cost-tolerance experimental data are used to set the training sets to establish a cost-tolerance network. Three representation modes of the cost-tolerance relationship are presented. First, the cost-tolerance relationship is derived from the grid points setting by the required tolerance accuracy. Second, a reasonable manufacturing cost of an unknown cost-tolerance experimental pair can be derived by the simulation of a cost-tolerance network. Third, an inference model based on a network's output is proposed to express the scope of the cost variation of various tolerances by means of a cost band. Comparison is also made with the high-order polynomial power function and exponential function cost-tolerance curves adopted by Yeo et al . Analytical results prove that the application of the cost-tolerance analysis model based on neural networks yields better performance in controlling the average fitting error than all conventional fitting models. The representation model using a cost band can identify precisely the possible cost variation range and reduce the chances of error in the tolerance design and cost estimation. It can thus provide important references for tolerance designers and process planners.     

Tolerance chart balancing for machining process planning

This paper provides a mathematical model for tolerance chart balancing for machining process planning. The formulation of the proposed model is based on the graphic ‘rooted tree’ representation technique in describing the sequence of the machining process. The criteria considered in the presented study are based on the combined effects of manufacturing cost and quality loss under constraints such as process capability limits, design functionality restrictions, and product quality requirements. Applications of these models include minimizing the total cost of manufacturing activities and quality related issues with process selection and tolerance allocation in machining process planning, particularly in the early stage of planning.     

Selection of management accounting systems in Just-In-Time and Theory of Constraints-based manufacturing

In today's competitive business environments, a firm's long-term survival rests heavily on its ability to sustain manufacturing superiority over its competitors. To provide the firm with detailed guidelines for sustaining manufacturing superiority, this paper examines the impact of different management accounting systems, manufacturing control systems and time horizon on manufacturing performance in an enterprise resource planning integrated environment. These management accounting systems include traditional costing, activity-based costing and throughput accounting. The manufacturing control systems include Just-In-Time- and Theory of Constraints-based manufacturing. Through a series of simulation experiments, it was found that activity-based costing provided higher short- and long-term profit, better customer service and lower work-in-process inventory than traditional costing and throughput accounting in situations where firms have high overhead costs and relatively low labour and material costs, while carrying ending inventories because of demand uncertainty. Traditional costing also outperformed throughput accounting by exploiting the real-time information sharing capabilities of an enterprise resource planning system. Just-In-Time manufacturing outperformed Theory of Constraints with respect to short- and long-term profitability, customer service, and work-in-process inventory because of differences in buffer inventory policies and sequencing rules. However, time horizon and its interaction with management accounting systems had no impact on the manufacturing performance. In addition, the results suggest that a management accounting system that depicts the manufacturing process tended to provide more accurate product cost information and resulted in a better system performance than the others.