Fuzzy Scheduling Model And Algorithm For Flexible Job Shop

Workshop scheduling is an important technology node for the digital transformation of manufacturing.Due to the increasingly complex production process of factory products and the increasing number of uncertain factors in flexible production shop,it becomes more and more difficult to obtain accurate production parameters.The flexible job shop scheduling problem with uncertain production parameters cannot be solved well by traditional scheduling methods.With the development of fuzzy set theory,the use of fuzzy numbers to represent the uncertain production parameters in flexible job shops can help improve the rationality and reliability of scheduling solutions.Therefore,how to find an effective method to solve such fuzzy flexible job shop scheduling problems has become an important research topic in academia and industry.In this paper,based on the particle swarm optimization and its important improvement algorithms,the single-objective and multi-objective fuzzy flexible job shop scheduling problems are studied as follows.(1)The definition,representation method of fuzzy sets and commonly used fuzzy numbers are introduced,and a defuzzification method for fuzzy numbers is proposed.For the flexible job shop scheduling problem with uncertain production parameters,the fuzzy numbers are used to represent production parameters that fluctuate within a certain range of values,and a fuzzy scheduling model of flexible job shop is established.The classification of the model is described and the mathematical description of the model constraints and the scheduling objective function is given.(2)For the single-objective fuzzy flexible job shop scheduling problem,the coding and decoding method of the problem and the discretization rules of the solution vector are given,and a hybrid adaptive particle swarm optimization is designed.The position update formula of the particles is improved by the Cauchy distribution,and the adaptive inertia weight is introduced to enhance the algorithm’s merit-seeking ability.The effectiveness of the two improved strategies is verified by benchmark test functions.In order to better solve the fuzzy scheduling problem,genetic crossover operations are performed on the process codes mapped by the excellent particles,and simulated annealing is introduced to further enhance the depthseeking ability of the algorithm.Finally,the superiority of the proposed algorithm is verified by simulation experiments.(3)For the multi-objective fuzzy flexible job shop scheduling problem,a multi-strategy fusion quantum particle swarm optimization with better multi-objective seeking capability is designed.The effectiveness of the standard algorithm for solving multi-objective optimization problems is verified using multi-objective test functions.On the basis of the standard algorithm,the chaotic mapping is used to initialize the population,and the lévy flight is introduced to enhance the algorithm’s ability to jump out of local optimum.A neighborhood search method based on machine mutation is designed for local search,and elite individual diversity is maintained using genetic crossover operators and simulated annealing.Finally,a multiindicator weighted grey target decision model is introduced to solve the multi-objective solution centralized scheduling scheme decision problem.The simulation experiments verify the superiority and effectiveness of the algorithm and decision model.