Research Of Scheduling Methods Based On Swarm Intelligence For Uncertainty Production Processes

With the development of science and technology, the production scale is growing and the complexity is increasing, both of which result in the increasingly fierce market competition. In the complex and changeable market environment, the competition of a corporation not only depends on advanced production technology and equipment, but also must relays on advanced production management. The production scheduling is the core for advanced manufacturing systems to develop management technology, planning, optimization technology, automation and computer technology. And the keys to implement advanced manufacturing and improve production efficiency are the effective scheduling method and the research of optimization technology. As a new intelligent algorithm, ant colony algorithm has gradually becoming the research interest because of its excellent characters such as good positive feedback, robustness, groups, parallelism and so on. This study is started with the application of ant colony algorithm in production scheduling. The work we have done as follows:On the basis of studying uncertainties in the production scheduling problem, we present the Job shop fuzzy scheduling model for the scheduling problem with fuzzy processing time and fuzzy due date. In the present model, fuzzy processing time and fuzzy due date are denoted by triangular fuzzy numbers and Trapezoid fuzzy numbers respectively, and the scheduling goal is the greatest satisfaction of average delivery. To address the fuzzy scheduling problem, we have improved the basic ant colony algorithm and present a new state transition rules, at the same time use adaptive pheromone updating strategy that can quickly get out of part convergence. The simulation results show that adaptive ant colony algorithm is effective to solve Job Shop fuzzy scheduling problem.For the fact that most of scheduling tasks are flexible and dynamic in the actual industrial production, the improved ant colony algorithm is used to solve practical scheduling problems. At last we have completed the development of intelligent scheduling system which combines theory and practice.