Using data mining to find patterns in genetic algorithm solutions to a job shop schedule

摘要: Abstract This paper presents a novel use of data mining algorithms for the extraction knowledge from large set job shop schedules. The purposes this work is to apply methodologies explore patterns in generated by genetic algorithm performing scheduling operation and develop rule scheduler which approximates algorithm's scheduler. Genetic are stochastic search based on mechanics genetics natural selection. Because inheritance, characteristics survivors after several generations should be similar. In using scheduling, solution an operational sequence resource allocation. Among these optimal or near solutions, similar relationships may exist between operations sequential order. An attribute-oriented induction methodology was used relationship operations’ its attributes rules has been developed. These can duplicate performance identical problem provide solutions that generally superior simple dispatching problems.