Optimisation of cutting parameters using a multi-objective genetic algorithm

This paper presents a new optimisation technique based on genetic algorithms (GA) for determination of cutting parameters in machining operations. The cutting parameters considered in this study are cutting speed, feed rate and cutting depth. The effect of these parameters on production time, production cost and roughness is mathematically formulated. A genetic algorithm with multiple fitness functions is proposed to solve the formulated problem. The proposed algorithm finds multiple solutions along the Pareto optimal frontier. Experimental results show that the proposed algorithm is both effective and efficient, and can be integrated into an intelligent process planning system for solving complex machining optimisation problems.     

Optimisation of a set of or principal components control charts using genetic algorithms

When a multivariate process is to be monitored, there are the options of employing a set of univariate control charts or a single multivariate chart. This paper shows how to effectively design a multivariate control scheme consisting of two or three X charts, using genetic algorithms to optimise the charts parameters. The procedure is implemented using software tools, which we designed. A complete performance comparison of the scheme with the Hotelling's T ² control chart can be made in order to help the user in choosing the most adequate option for the process under consideration. Also, if the user prefers to employ charts based on principal components rather than on the original variables, the software can be used in the same way to optimise a set of two or three control charts based on these components, and to compare their performance with the performance of the T ² chart on the principal components.     

Batch sizes optimisation by means of queueing network decomposition and genetic algorithm

Batch sizes have a considerable impact on the performance of a manufacturing process. Determining optimal values for batch sizes helps to reduce inventories/costs and lead times. The deterministic nature of the available batch size optimisation models reduces the practical value of the obtained solutions. Other models focus only on critical parts of the system (e.g., the bottleneck). In this paper, we present an approach that overcomes important limitations of such simplified solutions. We describe a combination of queueing network analysis and a genetic algorithm that allows us to take into account the real characteristics of the system when benefiting from an efficient optimisation mechanism. We are able to demonstrate that the application of our approach on a real-sized problem with 49 products allows us to obtain a solution (values for batch sizes) with less than 4% relative deviation of the cycle time from the exact minimal value.     

Development of a post-flood road maintenance strategy: case study Queensland,Australia

Currently, no road authority takes into account flooding in road deterioration (RD) models; as a result, post-flood rehabilitation treatments may be sub-optimal. This paper proposes a new approach to the development of a post-flood maintenance strategy. The recently developed roughness and rutting-based RD models with flooding, by the current authors, are used as input to predict pavement deterioration after a flood (i.e. assuming a flood in year 1). The HDM-4 model has been used to get the post-flood maintenance strategy with constrained and unconstrained budget, where post-flood rehabilitation starts from year 2. The road groups in state road network of Queensland, Australia, are used as the case study. The unconstrained budget solution aims to keep the network in an excellent condition at a cost of $49.7bn with the possible strongest treatments. The constrained budget strategy uses agency cost and pavement performance as constraints in optimisation and provides a reasonable solution. This strategy requires about $26.1bn in life cycle, which is close to the main road authority of Queensland’s post-flood rehabilitation programme. The paper discusses two other strategies on maximise economic benefits and budget optimisation. It is expected that a road authority would properly investigate its flood-damaged roads before implementation. The paper shows pavement performances with the post-flood strategy. The need for a RD model to predict deterioration after a flood and for post-flood treatment selection is also highlighted.     

An adaptive genetic algorithm for large-size open stack problems

The problem of minimising the maximum number of open stacks arises in many contexts (production planning, cutting environments, very-large-scale-integration circuit design, etc.) and consists of finding a sequence of tasks (products, cutting patterns, circuit gates, etc.) that determines an efficient utilisation of resources (stacks). We propose a genetic approach that combines classical genetic operators (selection, order crossover and pairwise interchange mutation) with an adaptive search strategy, where intensification and diversification phases are obtained by neighbourhood search and by a composite and dynamic fitness function that suitably modifies the search landscape. Computational tests on random and real-world benchmarks show that the proposed approach is competitive with the state of the art for large-size problems, providing better results for some classes of instances.     

Optimisation of ISI interval using genetic algorithms for risk informed in-service inspection

Risk Informed In-Service Inspection (RI-ISI) aims at prioritising the components for inspection within the permissible risk level thereby avoiding unnecessary inspections. Various constraints such as risk level, radiation exposure to the workers and cost of inspections are encountered, while planning the inspection programme. This problem has been attempted to solve using genetic algorithms, which has already established its suitability in optimizing Surveillance and Maintenance activities in Nuclear Power Plants. The paper describes the application of genetic algorithm in optimizing the ISI of feeders, which are large in number and also fall in the same inspection category.     

Physical Internet,conventional and hybrid logistic systems: a routing optimisation-based comparison using the Eastern Canada road network case study

The Physical Internet (PI) logistics system is an innovative logistics concept that has been gathering a lot of attention lately. This system consists of open, modular and shared containers and transit hubs to move goods globally. The purpose of this paper is to compare the performance of PI with regard to the conventional (CO) logistics system in order to quantify the advantages and disadvantages of PI from a truck and driver routing perspective with an explicit constraint on maximum return time for drivers. The comparison presented in this work is carried out through Monte-Carlo simulation within a sequential three-phase optimisation framework. Based on our analysis, PI reduces driving distance (and time), GHG (greenhouse gas) emissions and the social cost of truck driving. On the other hand, it increases the number of container transfers within the PI logistics centres. This insight is a contribution of the paper and reinforces the current literature on PI. The other main contribution of the paper is a validation of the claim that the number of drivers who can go back home at the end of a work day remains consistently high in PI, regardless of the traffic level.     

Hybridizing a genetic algorithm with an artificial immune system for global optimization

This paper proposes an algorithm based on a model of the immune system to handle constraints of all types (linear, nonlinear, equality, and inequality) in a genetic algorithm used for global optimization. The approach is implemented both in serial and parallel forms, and it is validated using several test functions taken from the specialized literature. Our results indicate that the proposed approach is highly competitive with respect to penalty-based techniques and with respect to other constraint-handling techniques which are considerably more complex to implement.     

A genetic algorithm and queuing theory based methodology for facilities layout problem

Facilities layout, being a significant contributor to manufacturing performance, has been studied many times over the past few decades. Existing studies are mainly based on material handling cost and have neglected several critical variations inherent in a manufacturing system. The static nature of available models has reduced the quality of the estimates of performance and led to not achieving an optimal layout. Using a queuing network model, an established tool to quantify the variations of a system and operational performance factors including work-in-process (WIP) and utilisation, can significantly help decision makers in solving a facilities layout problem. The queuing model utilised in this paper is our extension to the existing models through incorporating concurrently several operational features: availability of raw material, alternate routing of parts, effectiveness of a maintenance facility, quality of products, availability of processing tools and material handling equipment. On the other hand, a queuing model is not an optimisation tool in itself. A genetic algorithm, an effective search process for exploring a large search space, has been selected and implemented to solve the layout problem modelled with queuing theory. This combination provides a unique opportunity to consider the stochastic variations while achieving a good layout. A layout problem with unequal area facilities is considered in this paper. A good layout solution is the one which minimises the following four parameters: WIP cost, material handling cost, deviation cost, and relocation cost. Observations from experimental analysis are also reported in this paper. Our proposed methodology demonstrates that it has a potential to integrate several related decision-making problems in a unified framework.     

基于遗传算法的三维无线传感器网络定位新算法

针对三维无线传感器网络节点自身定位问题,提出了一种基于遗传算法的新定位算法。该算法通过分析未知节点与它的无线射程范围内的已知节点之间的通讯约束和距离测量,对未知节点建立数学模型;针对此数学模型利用遗传算法求解,把该解作为未知节点的估计位置。理论分析和试验结果表明,该算法具有很强的健壮性,未知节点的失效和新节点的加入不会影响算法的性能,并且算法定位精度高,条件简单,适合各种规模的无线传感器网络的节点定位。     

遗传算法在两分频扬声器系统设计中的应用

文章论述了运用遗传算法优化设计两分频扬声器分频网络的方法。使用实测的扬声器阻抗、频响数据,模拟扬声器接入分频网络后扬声器系统的响应,并运用遗传算法对系统响应进行优化设计。文中给出的实例表明,设计结果和实测的曲线是一致的。     

A method combining rules with genetic algorithm for minimizing makespan on a batch processing machine with preventive maintenance

This paper considers the problem of minimising makespan on a single batch processing machine with flexible periodic preventive maintenance. This problem combines two sub-problems, scheduling on a batch processing machine with jobs’ release dates considered and arranging the preventive maintenance activities on a batch processing machine. The preventive maintenance activities are flexible but the maximum continuous working time of the machine, which is allowed, is determined. A mathematical model for integrating flexible periodic preventive maintenance into batch processing machine problem is proposed, in which the grouping of jobs with incompatible job families, the starting time of batches and the preventive maintenance activities are optimised simultaneously. A method combining rules with the genetic algorithm is proposed to solve this model, in which a batching rule is proposed to group jobs with incompatible job families into batches and a modified genetic algorithm is proposed to schedule batches and arrange preventive maintenance activities. The computational results indicate the method is effective under practical problem sizes. In addition, the influences of jobs’ parameters on the performance of the method are analyzed, such as the number of jobs, the number of job families, jobs’ processing time and jobs’ release time.     

An improved multi-objective genetic algorithm for heterogeneous coverage RFID network planning

Recent research has demonstrated the potential benefits of radio frequency identification (RFID) technology in the supply chain and production management via its item-level visibility. However, the RFID coverage performance is largely impacted by the surrounding environment and potential collisions between the RFID devices. Thus, through RFID network planning (RNP) to achieve the desired coverage within the budget becomes a key factor for success. In this study, we establish a novel and generic multi-objective RNP model by simultaneously optimising two conflicted objectives with satisfying the heterogeneous coverage requirements. Then, we design an improved multi-objective genetic algorithm (IMOGA) integrating a divide-and-conquer greedy heuristic algorithm to solve the model. We further construct a number of computational cases abstracted from an automobile mixed-model assembly line to illustrate how the proposed model and algorithm are applied in a real RNP application. The results show that the proposed IMOGA achieves highly competitive solutions compared with Pareto optimal solutions and the solutions given by four recently developed well-known multi-objective evolutionary and swarm-based optimisers (SPEA2, NSGA-II, MOPSO and MOPS²O) in terms of solution quality and computational robustness.     

A guided search genetic algorithm using mined rules for optimal affective product design

Affective design is an important aspect of new product development, especially for consumer products, to achieve a competitive edge in the marketplace. It can help companies to develop new products that can better satisfy the emotional needs of customers. However, product designers usually encounter difficulties in determining the optimal settings of the design attributes for affective design. In this article, a novel guided search genetic algorithm (GA) approach is proposed to determine the optimal design attribute settings for affective design. The optimization model formulated based on the proposed approach applied constraints and guided search operators, which were formulated based on mined rules, to guide the GA search and to achieve desirable solutions. A case study on the affective design of mobile phones was conducted to illustrate the proposed approach and validate its effectiveness. Validation tests were conducted, and the results show that the guided search GA approach outperforms the GA approach without the guided search strategy in terms of GA convergence and computational time. In addition, the guided search optimization model is capable of improving GA to generate good solutions for affective design.     

Unequal-area stochastic facility layout problems: solutions using improved covariance matrix adaptation evolution strategy,particle swarm optimisation,and genetic algorithm

Determining the locations of departments or machines in a shop floor is classified as a facility layout problem. This article studies unequal-area stochastic facility layout problems where the shapes of departments are fixed during the iteration of an algorithm and the product demands are stochastic with a known variance and expected value. These problems are non-deterministic polynomial-time hard and very complex, thus meta-heuristic algorithms and evolution strategies are needed to solve them. In this paper, an improved covariance matrix adaptation evolution strategy (CMA ES) was developed and its results were compared with those of two improved meta-heuristic algorithms (i.e. improved particle swarm optimisation [PSO] and genetic algorithm [GA]). In the three proposed algorithms, the swapping method and two local search techniques which altered the positions of departments were used to avoid local optima and to improve the quality of solutions for the problems. A real case and two problem instances were introduced to test the proposed algorithms. The results showed that the proposed CMA ES has found better layouts in contrast to the proposed PSO and GA.     

Bayesian spatial joint modeling of traffic crashes on an urban road network

This study proposes a Bayesian spatial joint model of crash prediction including both road segments and intersections located in an urban road network, through which the spatial correlations between heterogeneous types of entities could be considered. A road network in Hillsborough, Florida, with crash, road, and traffic characteristics data for a three-year period was selected in order to compare the proposed joint model with three site-level crash prediction models, that is, the Poisson, negative binomial (NB), and conditional autoregressive (CAR) models. According to the results, the CAR and Joint models outperform the Poisson and NB models in terms of model fitting and predictive performance, which indicates the reasonableness of considering cross-entity spatial correlations. Although the goodness-of-fit and predictive performance of the CAR and Joint models are equivalent in this case study, spatial correlations between segments and the connected intersections are found to be more significant than those solely between segments or between intersections, which supports the employment of the Joint model as an alternative in road-network-level safety modeling.     

A risk-based optimisation methodology for pavement management system of county roads

A pavement management system (PMS) is a strategic and systematic process to maintain and upgrade the road network. When funding is limited, it is very important to identify the best mix of road preservation projects that provides the maximum benefits to society in terms of overall life cycle cost of the road network. The most common factors that play an important role for identifying projects are the following: budget, traffic volume, Present Serviceability Index (PSI) and risk associated with selecting treatment types. This research develops an optimisation methodology for county paved roads that identify the best mix of preservation projects within budget, maximising traffic (passengers and trucks traffic) on treated roads, maximising the weighted average PSI, and minimising the risk. This methodology will facilitate a statewide implementation of PMS for counties in the state of Wyoming.     

基于神经网络遗传算法的GFRP拉挤双目标优化

为研究玻璃钢(GFRP)拉挤工艺参数对复合材料性能的影响,优化最佳拉挤工艺参数,建立了拉挤工艺过程数学模型,结合基于有限元/有限差分的间接解耦法进行求解,模拟得到了拉挤过程中GFRP内部的非稳态温度场和固化度变化情况.分别采用布拉格光栅光纤温度传感器和索氏萃取法检测拉挤GFRP内部的温度与固化度,实测温度和固化度均与模拟温度和固化度吻合,验证了数值模拟程序的正确性.以数值模拟结果为样本,建立反向传播神经网络,得到拉挤工艺参数(固化温度、拉挤速度)与GFRP固化度之间的非线性相关关系,再结合遗传算法解决拉挤过程中固化炉温度和拉挤速度双目标优化问题.优化得到的拉挤工艺参数可在保证复合材料固化度达标的情况下,提高拉挤速度降低固化炉温度,优化效果显著.神经网络遗传算法优化方法能有效解决此类具有复杂非线性关系的多目标优化问题.     

基于遗传算法的神经网络被动声呐目标分类研究

被动声呐目标识别系统中目标分类器的设计和训练是一项重要内容，本文设计了目标分类器的神经网络结构，提出了一种用改进的遗传算法训练神经网络分类器的新方法，最后，对海上实录的Ａ，Ｂ，Ｃ三类目标噪声进行了分类识别，实验结果表明基于遗传算法的神经网络分类器比传统的基于ＢＰ算法的神经网络分类源泛化性能有明显提高。     

Analysis of closed loop supply chain using genetic algorithm and particle swarm optimisation

This paper describes a relational database system for semi-generative process planning for sheet metal parts that emulates expert system capabilities. The system integrates a feature-based relational database for the parts, a forward chaining rule-based strategy for machine selection, both global and feature-specific execution of the rules and a graph theoretic cost optimization model for optimal process plan selection. This system, which is currently being developed for a sheet metal fabrication company, suggests that, using the experience of shopfloor personnel, an efficient integration of feature-based process planning and expert system strategies can be accomplished.