Dynamic Job Shop Scheduling Problem With New Job Arrivals Using Hybrid Genetic Algorithm

The present paper tackles the dynamic job shop scheduling problem (DJSSP), aiming to schedule a new set of jobs while minimizing the completion time of all operations. The problem is an NP-hard combinatorial optimization problem. This contribution proposes an optimal scheduling method based on the evolutionary genetic algorithm approach. The difficulty of this problem is to comprehensively find the best direction of a candidate solution while maintaining the minimum total completion time known as the makespan and denoted as Cmax. To adapt the system to changes and perform the scheduling of a new job, a local search could be an appropriate solution to fix and repair the problem by guiding the search directions following the job's arrival. Experiment-based statistical analysis shows that the proposed model has better convergence and accuracy than state-of-the-art algorithms.