Fuzzy Job_Shop Scheduling Based On Chaotic Crow Search Algorithm

Job shop scheduling,as a key link in guiding production operations,makes an important impact in the management of manufacturing enterprise.A good scheduling solution can improve production efficiency and reduce production costs.Fuzzy job shop scheduling problem(FJSSP)is proposed by introducing fuzzy conditions based on the job shop scheduling problem(JSSP),which is more suitable for actual production.This paper mainly studied the FJSSP with fuzzy processing time and the processing time is expressed by triangular fuzzy number(TFN),established the mathematical model of FJSSP,and proposed the solution methods.The main work of this paper is as follows:Firstly,the chaotic crow search algorithm(CCSA)was used to solve the FJSSP.The process-based coding method was used to encode and decode the solution of the FJSSP.At the same time,a repair method was designed to make the solution that calculated by the algorithm satisfies the encoding rules and constraints.Secondly,since the CCSA search method that a crow individual follows the optimal solution searched by another individual in the crow group is too singular.In order to enhance the individual’s ability in neighborhood search when follows target solution,this paper introduces a mutation operator to enrich the individual’s search behavior.In order to enhance the search efficiency of the algorithm,so that the individual can follows to the high-quality solution in the group while ensuring the population diversity to avoid prematurely falling into the local optimal solution,this paper proposed a multiple optimal individual set based on the cosine similarity,the optimal individual as the target to guide the search,and at the same time the individuals with lower similarity in the group as the target to ensure the diversity of the population information.We tested and compared the proposed algorithm on eight typical examples.The result verified that the proposed algorithm has good solving quality,speed and stability,and it also proved that the improvements can improve the search ability and convergence ability of the algorithm effectively.Because the solution space of FJSSP is large and complex,Especially large scale FJSSP,the search strategy of CCSA seems to be weak in solving such problems.In order to improve the search ability of the algorithm further,this paper analyzed the scheduling problem based on Gantt chart model,and a search method based on the reduction of machine’s spare time was proposed.At the same time,a coding model suitable for this method and a method of mutual conversion with process-based coding were designed.Then,the method was integrated into the CCSA to form a new CCSA with hybrid search strategy.Finally,five typical examples were selected for comparison test,which verified the improvements of the proposed algorithm in solving quality and stability.