An improved genetic algorithm for the flexible job shop scheduling problem with multiple time constraints

The flexible job shop scheduling problem is a very important problem in factory scheduling. Most of existing researches only consider the processing time of each operation, however, jobs often require transporting to another machine for the next operation while machines often require setup to process the next job. In addition, the times associated with these steps increase the complexity of this problem. In this paper, the flexible job scheduling problem is solved that incorporates not only processing time but setup time and transportation time as well. After presenting the problem, an improved genetic algorithm is proposed to solve the problem, with the aim of minimizing the makespan time, minimizing total setup time, and minimizing total transportation time. In the improved genetic algorithm, initial solutions are generated through three different methods to improve the quality and diversity of the initial population. Then, a crossover method with artificial pairing is adopted to preserve good solutions and improve poor solutions effectively. In addition, an adaptive weight mechanism is applied to alter mutation probability and search ranges dynamically for individuals in the population. By a series of experiments with standard datasets, we demonstrate the validity of our approach and its strong performance.