Genetic Algorithms for Solving Bicriteria Dynamic Job Shop Scheduling Problems with Alternative Routes

Abstract

Solving scheduling problems with a single criterion is considered unsatisfactory for real-world applications. Therefore, more attention has been given to multiple objective scheduling problems. In this paper, we use a genetic algorithms to solve job shop scheduling problems with alternative routes and dynamic job arrival in order to simultaneously minimize the maximum lateness and makespan. Firstly, genetic algorithms have been applied to find a set of optimum feasible solutions for the makespan criterion. Individuals or solutions with values less than or equal to the value of maximum lateness with minimum makespan are then used to form the initial population in genetic algorithms for the second criterion in order to minimize the maximum lateness. A method of finding non-dominated solutions is then proposed, and weighted-sum is used to find the most desirable solution based on the weight of each criteria. Finally the model is tested using different instances, with the obtained results demonstrating the effectiveness of the proposed method to solve bicriteria dynamic job shop scheduling problems with alternative routes.