An enhanced single objective genetic algorithm scheduling tool for solving very large scheduling problems in the capital goods industry

In this paper, an Enhanced Single Objective Genetic Algorithm Scheduling Tool (ESOGAST) is presented that is capable of solving very large scheduling problems such as those encountered in the capital goods industry. The tool minimises the penalties caused by the early or late delivery of components, assemblies and final products. The tool is optimised for speed; it runs more than 5000 times faster than the tool developed by Pongcharoen et al [1]. It is therefore capable of solving much larger problems within a reasonable amount of time. The ESOGAST includes an enhanced repair process to optimise the performance of the tool for scheduling the production of very complex products with many levels of product structure under finite capacity conditions. This paper describes the Genetic Algorithm and the data structures used in detail. A case study that used data obtained from a collaborating capital goods company is presented. A series of experiments that were used to identify the best combination of genetic operators and parameter settings is described. The work optimised an 18 month schedule for a manufacturing facility with 52 machines that produced several families of complex products with deep product structure.

[1]  Hisao Ishibuchi,et al.  Performance Evaluation of Genetic Operations for Flowshop Scheduling , 1996 .

[2]  L. Darrell Whitley,et al.  Scheduling Problems and Traveling Salesmen: The Genetic Edge Recombination Operator , 1989, International Conference on Genetic Algorithms.

[3]  D. J. Smith,et al.  A Study of Permutation Crossover Operators on the Traveling Salesman Problem , 1987, ICGA.

[4]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[5]  Pupong Pongcharoen,et al.  The development of genetic algorithms for the finite capacity scheduling of complex products, with multiple levels of product structure , 2004, Eur. J. Oper. Res..

[6]  Shirley Coleman,et al.  Overcoming complexity via statistical thinking: optimising genetic algorithms for use in complex scheduling problems via designed experiments , 2002 .

[7]  John J. Grefenstette,et al.  Optimization of Control Parameters for Genetic Algorithms , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[8]  K. Endebrock,et al.  Design for recyclability of capital goods , 1999, Proceedings First International Symposium on Environmentally Conscious Design and Inverse Manufacturing.

[9]  Kalyan Ghosh,et al.  A simplified and efficient representation for evaluation and selection of assembly sequences , 2003, Comput. Ind..

[10]  Goldberg,et al.  Genetic algorithms , 1993, Robust Control Systems with Genetic Algorithms.

[11]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[12]  G. Syswerda,et al.  Schedule Optimization Using Genetic Algorithms , 1991 .

[13]  Heinz Mühlenbein,et al.  Evolution algorithms in combinatorial optimization , 1988, Parallel Comput..

[14]  Pm Braiden,et al.  Computer-aided production management issues in the engineer-to-order production of complex capital goods explored using a simulation approach , 2000 .

[15]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1992, Artificial Intelligence.

[16]  Lawrence Davis,et al.  Job Shop Scheduling with Genetic Algorithms , 1985, ICGA.

[17]  Brian W. Kernighan,et al.  The C Programming Language , 1978 .

[18]  P. Pongcharoena,et al.  Determining optimum Genetic Algorithm parameters for scheduling the manufacturing and assembly of complex products , 2002 .