Discrete Manufacturing Production Shop Scheduling Optimization

Job-shop scheduling, a high simplification of practical production, is a coexistent problem in distribute manufacturing industry. Solving the problem will be the key to make the full use of resources, work out the rational production plan of workshop and enterprise, raise the satisfaction of customers, enhence the competition ability of enterprise. Job-shop scheduling is a kind of hard problem about combinatorial optimization. It is a hot point in the manufacturing researching. Many efforts have been made for several decades in order to solve the hard problem and have made considerable achievement. But a large amount of work must be done if then problem is solved completely. Several problems of this field in depth have been studied and some good results have been obtained in this dissertation.1. The definition of distribute manufacturing industry scheduling problem is given. Its characteristics are analyzed. The possibility and complexity of computing it are discussed. And the main techniques and methods of solving it are reviewed.2. On the base of deep research about basic principles of genetic algorithm, an advanced genetic algorithm for Job-shop scheduling problem is presented. In the algorithm, the encoding'method based on number of work is given. This encoding method avoided the illegal solution of Job-shop scheduling problem, and increased the search efficiency of GA.3. GA was combined with TOC (theories of constraints) to solve Job-shop scheduling problem. In the process of optimization, the rule of best efficiency of bottleneck was prior satisfied, then, the second time optimization was conducted and then. The rule of best efficiency of bottleneck ensures the space of searching of GA was limited into smaller scope. It accelerates the speed of GA' s constringency.4. Studied the familiar abrupt affairs in the production process, and classified the affairs by its influence to production process property. Andprovided correspond replay strategies and processes for each kind of affair.5. Realized simulation of algorithms presented in this dissertation based on VB6.0. The algorithms are evaluated and compared by benchmark problems. The results of computer simulation experiment indicate that the algorithms are applicable and effective.In the end, further developments on distribute production factory scheduling are explored.