Applied Soft Computing 13 (2013) 1329–1331
Contents lists available at SciVerse ScienceDirect
Applied Soft Computing journal homepage: www.elsevier.com/locate/asoc
Editorial
Special issue on hybrid evolutionary systems for manufacturing processes
1. Introduction From past to present, classical evolutionary algorithms, such as genetic algorithms, genetic programming, evolutionary programming and evolution strategy, have contributed to optimize a wide range of manufacturing processes, whose demands to be more robust, more flexible, more responsive, more complex and more efficient are ever increasing. In today’s competitive market, manufacturing process problems have to be solved with impeccable quality in short computational time. In general, reasonable results regarding particular manufacturing processes can be obtained by applying classical evolutionary algorithms which may not achieve the most convincing solution with the highest quality for a particular manufacturing process. To achieve the highest quality solution for a particular manufacturing process, integration of special techniques into the classical evolutionary algorithm is usually required. The resulting systems are called hybrid evolutionary systems. Usually hybrid evolutionary systems are integrated with classical optimization methods, heuristic algorithms or other computational intelligence methods in order to enhance the effectiveness of the classical evolutionary algorithms. Literature of evolutionary computation shows that the hybrid evolutionary systems are usually able to obtain higher quality solutions with smaller computational time than those obtained by the classical evolutionary algorithms for a particular manufacturing process. Although the application oriented research with hybrid evolutionary systems has currently reached an impressive state, there are a number of critical issues regarding the design of hybrid evolutionary systems with the rapidly growing complexity of manufacturing processes and demanding manufacturing qualities. This special issue aims to bring together researchers from academia and industry to report and review the latest progress in application oriented research with hybrid evolutionary systems, to explore new applications in manufacturing processes, to design new hybrid evolutionary systems for solving specific problems in manufacturing processes and finally to create awareness of hybrid evolutionary systems for a wider audience of practitioners. Target authors included research students, researchers and scientists from computational intelligence, manufacturing or product design engineers involved high precisions and high quality design in manufacturing processes or product design manufacturing or product design professionals The following areas of hybrid evolutionary systems for manufacturing processes were considered: 1) Developing hybrid evolutionary systems for solving real world problems in manufacturing processes. 1568-4946/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.asoc.2013.01.001
2) Designing hybrid evolutionary systems by co-operating evolutionary algorithms with other modern optimization algorithms (such as particle swarm optimization, simulated annealing, artificial immune system, ant colony etc.). 3) Modifying evolutionary operations by integrating with computational intelligence techniques (such as fuzzy systems, neural networks, support vector machines etc.) so as to produce an effective hybrid evolutionary system. 4) Enhancing evolutionary algorithms by co-operating with classical optimization or statistical methods. 5) Using hybrid evolutionary systems for handling constrained, multi-objective and large scale optimization problems in manufacturing processes. 6) Applying hybrid evolutionary systems in real manufacturing processes such as scheduling, allocation etc. 7) Parallel hybrid evolutionary systems for practical applications in manufacturing processes. 8) Using hybrid evolutionary systems for solving time-varying optimization problems in manufacturing processes. 2. The papers in the special issue We received 65 high quality papers which have been considered for this special issue. After the strict review process, 23 papers were selected to be included in this special issue. They can be classified by two categories which are involved with combinatorial problems and parametrical problems for manufacturing processes. The 1st to 13th papers contribute on solving combinatorial problems on manufacturing processes, while the 15th to 23rd papers contribute on those for parametrical problems. 2.1. Combinatorial problems for manufacturing processes • The 1st paper titled ‘Heuristic-based Neural Networks for Stochastic Dynamic Lot-sizing Problems’ presents a hybrid algorithm integrated with genetic algorithm (GA) and bee algorithm for developing neural network which is used to solve multiperiod single-item lot sizing problem. • The 2nd paper titled ‘An AIS-based Hybrid Algorithm with PDRs for Multi-objective Dynamic Online Job Shop Scheduling Problem’ presents a hybrid method integrated with artificial immune systems and priority dispatching rules in order to solve the dynamic online job shop scheduling problem that new jobs continuously arrive at the job shop in a stochastic manner. Experiments demonstrate the efficiency and competitiveness of this hybrid method.
1330
Editorial / Applied Soft Computing 13 (2013) 1329–1331
• The 3rd paper titled ‘Glass Container Production Scheduling through Hybrid Multi-population based Evolutionary Algorithm’ presents a hybrid algorithm which is combined with a genetic algorithm, a simulated annealing, and a tailor-made heuristic. The hybrid algorithm is used to schedule the lots in the most cost-effective manner on a set of parallel molding machines that are fed by a furnace for melting glass. • The 4th paper titled ‘A Model Induced Max-Min Ant Colony Optimization for Asymmetric Traveling Salesman Problem’ presents a model induced max-min ant colony optimization for solving asymmetric traveling salesman problems. Results show that the proposed method outperforms classical hybrid metaheuristic algorithms. • The 5th paper titled ‘A Hybrid Intelligent Model for Order Allocation Planning in Make-to-order Manufacturing’ presents a novel hybrid optimization model, integrating a multi-objective memetic optimization process, a Monte Carlo simulation technique and a heuristic pruning technique, in order to tackle multiobjective order allocation planning problems involved various real-world production features. The effectiveness of the proposed method is evaluated by conducting industrial case studies. • The 6th paper titled ‘A Comparison of Hybrid Genetic Algorithm and Hybrid Particle Swarm Optimization to Minimize Makespan for Assembly Job Shop’ presents a hybrid GA mixed with POS to solve assembly job shop scheduling problem with lot streaming techniques. Results show that the hybrid GA can significantly outperform some tested hybrid PSO for solving the test problems under various system conditions. • The 7th paper titled ‘Inventory Based Two-objective Job Shop Scheduling Model and Its Hybrid Genetic Algorithm’ presents a hybrid GA integrated with a new crossover based on the critical path and a local search operator. Simulations involved with a set of benchmark problems show the effectiveness of the hybrid GA. • The 8th paper titled ‘Better Manufacturing Process Organization using Multi-agent Self-organization and Co-evolutionary Classifier Systems: The Multibar Problem’ presents a GA for solving a complex discoordination problem devised to test enhanced complexity for multi-agent systems. Results demonstrate the need to use the GA for better adaptation, rather than just a reinforcement learning algorithm, proving wrong the previous hypothesis. • The 9th paper titled ‘Self-optimization Module for Scheduling using Case-based Reasoning’ presents a learning module proposed for an autonomous parameterization of metaheuristics, integrated on a multi-agent system for the resolution of dynamic scheduling problems. A computational study is presented where the proposed module is evaluated, and obtained results are compared with previous ones. • The 10th paper titled ‘Hybrid Taguchi-differential Evolution Algorithm for Optimization of Multi-pass Turning Operations’ presents a hybrid optimization approach based on differential evolution algorithm and Taguchi’s method. The proposed approach is applied to two case studies for multi-pass turning operations in order to show its effectiveness in machining operations. Better results can be obtained. • The 11th paper titled ‘Hybrid Evolutionary Algorithm for Job Scheduling under Machine Maintenance’ presents Hybrid Evolutionary Algorithm for solving Job Scheduling Problems. The algorithm has been tested by solving a number of benchmark problems with comparison of the existing algorithms. The experimental results provide better understanding of job scheduling and necessary rescheduling under process interruption. • The 12th paper titled ‘A Hybrid Differential Evolution Algorithm for Job Shop Scheduling Problems with Expected Total Tardiness Criterion’ presents a hybrid differential evolution algorithm for the job shop scheduling problem with random processing times under the objective of minimizing the expected total tardiness.
Results show the effectiveness and efficiency of the proposed approach on solving different-scale test problems. • The 13th paper titled ‘A Hybrid Discrete Artificial Bee Colony Algorithm for Permutation Flowshop Scheduling Problem’ presents a hybrid bee colony algorithm to minimize the makespan in permutation flowshop scheduling problems. The efficiency of the proposed algorithm is tested by solving scheduling benchmarks. 2.2. Parametrical problems for manufacturing processes • The 14th paper titled ‘A Comparative Study of Different Local Search Application Strategies in Hybrid Metaheuristics’ presents comparison results on using different combinations of metaheusitic algorithms for solving process design problems. Effectiveness for using different combinations is indicated. • The 15th paper titled ‘A Flexible Algorithm for Fault Diagnosis in a Centrifugal Pump with Corrupted Data and Noise Based on ANN and Support Vector Machine Hyper-parameters Optimization’ presents a unique flexible algorithm to classify the condition of centrifugal pump based on support vector machine hyperparameters optimization and artificial neural networks. • The 16th paper titled ‘A Hybrid Descent Method with Genetic Algorithm for Microphone Array Placement Design’ presents a hybrid descent method consists of a genetic algorithm and a gradient-based method. This hybrid method has the descent property and helps to find the optimal placement for better beamformers. • The 17th paper titled ‘A New Iterative Mutually-coupled Hybrid GA–PSO Approach for Curve Fitting in Manufacturing’ presents a hybrid approach for B-spline curve reconstruction comprised of two bio-inspired techniques: GA and PSO. Experimental results show that the approach can reconstruct with very high accuracy extremely complicated shapes, and is unfeasible for reconstruction with current methods. • The 18th paper titled ‘Optimal Design of Laser Solid Freeform Fabrication System and Real-time Prediction of Melt Pool Geometry using Intelligent Evolutionary Algorithms’ presents a self-organizing pareto GA for simultaneous optimization of clad height and melt pool depth in laser solid freeform fabrication system process. Results show the effectiveness of GA in optimization of laser clad processes. • The 19th paper titled ‘Change Point Determination for a Multivariate Process using a Two-stage Hybrid Scheme’ presents a hybrid scheme involved logistic regression, multivariate adaptive regression splines model, support vector machine and change point identification strategy. Results reveal that the proposed hybrid scheme can effectively identify the change points of multivariate process. • The 20th paper titled ‘A Hybrid Approach based on Differential Evolution and Tissue Membrane Systems for Solving Constrained Manufacturing Parameter Optimization Problems’ presents a hybrid approach based on differential evolution algorithms and tissue-p Systems, which is used for solving a class of constrained manufacturing parameter optimization problems. Results show that the hybrid approach is superior or competitive to 17 optimization algorithms recently reported in the literature. • The 21st paper titled ‘Modeling and Prediction of Machining Quality in CNC Turning Process using Intelligent Hybrid Decision Making Tools’ presents a hybrid model based on neural networks trained by particle swarm optimization. Results show that the hybrid model can be used in automotive industries to decide machining parameters which attain quality with minimum power consumption and maximum productivity. • The 22nd paper titled ‘Forecasting the Yield of a Semiconductor Product with a Collaborative Intelligence Approach’ presents a
Editorial / Applied Soft Computing 13 (2013) 1329–1331
collaborative intelligence approach to improve the performance of semiconductor yield forecasting. By applying the collaborative intelligence approach to a dynamic random access memory case, the advantages of the proposed methodology over some existing approaches can be indicated. • The 23rd paper titled ‘A New Hybrid Differential Evolution Algorithm for the Selection of Optimal Machining Parameters in Milling Operations’ presents a novel hybrid optimization approach based on differential evolution algorithm and receptor editing property of immune system. The effectiveness of the proposed hybrid approach is compared with several intelligence algorithms. Special issue guest editors: Dr. Kit Yan Chan, Department of Electrical and Computer Engineering, Curtin University, WA, Australia Prof. Tharam Dillon, Department of Computer Science and Computer Engineering, La Trobe University, Australia
1331
Dr. Hak Keung Lam, School of Natural and Mathematical Sciences, King’s College London, U.K. Dr. Steve S.H. Ling, Faculty of Engineering and Information Technology, University of Technology, Sydney, Australia Prof. Hung T. Nguyen, Faculty of Engineering and Information Technology, University of Technology, Sydney, Australia Acknowledgments Finally, the guest editors of this special issue would like to thank Prof. Rajkumar Roy, Editor of the Applied Soft Computing journal, for providing us with the opportunity to edit this special issue. Thanks to Ms. Kalpana Balaraman and Ms. JoJo Xu for providing the assistance for this special issue on time. We would also like to thank the authors for submitting their valuable research outcomes as well as the reviewers who have critically evaluated the papers. We sincerely hope that readers will find this special issue very useful.