The Crowding Mechanism Genetic Algorithm for Solving Job Scheduling Problem

Considering that different types of jobs are required to be assigned to corresponding types of machines at a given time, job scheduling problem is solved to find the optimal processing sequences with two optimization objectives, i.e., minimal required time and lowest cost of consumption. The mathematical optimization model of job scheduling problem is established, and a crowding mechanism genetic algorithm (CMGA) is proposed to solve the model. The normalization method and the preference weight coefficient are used to deal with the optimization objectives, so as to determine the fitness function of the algorithm. Meanwhile, in order to find a better feasible solution that is hidden around the unfeasible solution, the penalty item is added to the fitness function to deal with the constraint conflict. Heuristic crossover is used to the growth of the optimal pattern of the individual, and selection based on the crowding mechanism is introduced to maintain the diversity of the population. Furthermore a local search process is adopted to perform searches for neighborhoods, and improve the quality of the solution. Finally, test data verify the correctness and validity of the proposed algorithm; and the proposed algorithm and the model have been applied to practical project.