Development of efficient genetic algorithm for open shop scheduling problem to minimise makespan

Abstract

Scheduling problem deals with the management of the resources in most optimal manner. In this research, open shop scheduling problem with an objective of minimising the makespan is considered. This problem comes under combinatorial category. Hence, development of an efficient heuristic is inevitable to minimise the makespan of the open shop scheduling problem. Meta-heuristic genetic algorithm (GA) is considered as it has the scope of improving performance measure of the problem. The performance of the genetic algorithm is influenced by selection method, crossover operator and mutation probability. Four different genetic algorithms are developed by varying selection method and crossover operator where three of this algorithm use newly proposed crossover operator while the fourth uses existing one-point crossover operator. A complete factorial experiment with three factors and three replications for each experimental combination is carried out on a set of problem instances with all the four genetic algorithm methods.