Genetic Algorithm with Parameters Optimization Mechanism for Hard Scheduling Problems

Genetic algorithms have demonstrated considerable success in providing efficient solutions to many non-polynomial-hard optimization problems. But unsuitable parameters may cause terrible solution for a specific scheduling problem. In this paper, we propose a genetic algorithm with parameters optimization mechanism, which can find the fittest control parameters, namely, number of population, probability of crossover, probability of mutation, for a given problem with a fraction of time, and then those parameters are used in the genetic algorithm for further more search operation to find optimal solution. For large scale problem, this novel genetic algorithm can get optimal control parameters effectively and get better solution, avoiding waste time caused by unfitted parameters. The algorithm is validated based on some benchmark problems of job shop scheduling.