Research For Intelligent Scheduling Decisions And System Development Based On DFS Of Elevator

In the information age, that enterprises build digital factory system is particularly important, and shop scheduling is a key link in the digital factory system. Thesis uses research and development of digital factory project of an elevator parts manufacturing enterprise as background, using firefly algorithm as the main measure of optimization, to shop intelligent scheduling technology doing research as follows:(1)Classification, characteristics, performance indicators, research methods of shop scheduling problem is explained more detailed. Based on this study the research status and the development trend of shop scheduling problem is illustrated;(2)For flexible job shop scheduling problem, a mathematical model of the problem is established, and thesis proposes the VEDFA to solve this problem.The algorithm discretizes standard firefly algorithm, using the step updated policy to update the position of fireflies, using virus infection operation to strengthen the collaboration and dynamic search capabilities between groups, using Interchange neighborhood structure to strengthen local search ability of the algorithm. And through examples the effectiveness of the algorithm is proved;(3)For shop scheduling problem of multi-objective with multi-resource constraining, a mathematical model of the problem is established.Thesis proposes a decoding algorithm of activity under multi-resource constraining, and proposes the IMOFA to solve this problem. The algorithm introduces a difference operator in the standard multi-objective firefly algorithm to strengthen collaboration and competition between groups, using the external file maintenance based on intensive distance to ensure uniformity of the solution set, enhanceing the algorithm of local search capabilities by introducing Baldwinian learning strategies. And thesis uses analytic hierarchy process and entropy theory for multi-objective to make decision. At last through an example the effectiveness of IMOFA and method of multi-objective decision is proved;(4)For dynamic scheduling problem of multi-objective with multi-resource constraining, thesis proposes re-scheduling policy based on scroll window of period and event-driven, and designs updated strategy of system status based on the freezing time to ensure the effective convergence between new scheduling scheme and the original scheduling scheme. On the basis a mathematical model of the dynamic scheduling problem is established.Thesis proposes the IMOFA-II to solve the re-scheduling scheme. The algorithm uses the "non-dominated sorting + crowding distance calculation" to compare firefly brightness, updating the position of fireflies by the way of crossover and mutation, designing the elitism strategy. Through an example the effectiveness of IMOFA-II is proved;Finally, based on the above theoretical research and the digital factory system, the database of intelligent scheduling system were developed. A prototype system were developed using Power Builder11.5 and MATLAB, realizating initially intelligent scheduling.