A genetic algorithm for scheduling of jobs on lines of press machines

Abstract

This paper introduces a Genetic Algorithm (GA) based solution technique for press machines scheduling problem of a car manufacturing factory. Firstly, the problem at hand, and the application of the GA in terms of coding, chromosome evaluation, crossover and mutation operators, are described in detail. After that, the GA is experimentally evaluated through some test problems. As the objective of the problem is the minimization of the completion time of the jobs, the GA based solution is compared with the Longest Processing Time (LPT) rule, and it is observed that the GA always produces better schedules than the LPT rule in a reasonably short amount of CPU time. © Springer-Verlag Berlin Heidelberg 2006.