Research On Job Shop Scheduling Problem Of The Elevator Car Production Based On Genetic Simulated Annealing Algorithm

JSP are not only an important issue in practical production of manufacturing system but also one of the difficulties of theoretic research. JSP is one of the key problems in the production management. Good algorithms for this problem can improve the production efficiency and product quality, reduce the cost, and so on. Consequently, it has been an attractive subject in academic and industrial communities.In this paper, based on the hardware factory production processing in Zhongshan, the scheduling strategy with the elevator car production was investigated. According to the elevator car production process, the mathematical model was formulated and was analyzed in detail. Aiming at characteristic of elevator car production scheduling problems, in order to avoid premature convergence and slow convergent speed at later evolution process of traditional genetic algorithm, a self-adapted hybrid genetic algorithm is proposed and analyzes design process of the algorithm detailedly. In view of the processing route of job having different, the traditional method of coding based on the processes is improved. The example is simulated by MATLAB and by comparing the improved staggered evolution coding method of genetic simulated annealing algorithm with the improved coding method of genetic algorithm and standard genetic simulated annealing algorithm; show that the scheduling model based on the improved staggered evolution coding method of genetic simulated annealing algorithm can get satisfying results.In the elevator car production process, different job have different processes, and the processing time are different on parallel machines that exist in every operations. So it not only needs to find the best sequence, but also to select the most appropriate machines in the elevator car production scheduling problems. We designed double coding based on the improved the method of coding, which arranged the sequences of jobs in the first layer, and the second is machine. Finally, the improved staggered evolution coding method of genetic simulated annealing algorithm is proposed to solve elevator car production scheduling problems, include simulated annealing operation, planning the genetic manipulation (selection, crossover, variation) and so on. Finally, according to the model of a bi-objective job shop scheduling, gives the scheme of elevator car production.