Research On The Scheduling Problem Of Flexible Work Workshop

Manufacturing is the foundation of the national economy,intelligent manufacturing is the key technical node of "Made in China 2025",and the research on flexible workshop scheduling is an important technical hotspot in the intelligent manufacturing process.With the emergence of advanced technology,the optimization of workshop scheduling is of increasing significance for enterprises to produce new products faster,reduce costs in the production process,and reduce machine losses.Based on the above background,this paper studies the scheduling problem of flexible work workshop.On the basis of systematically expounding the research background,research status and relevant theoretical knowledge of flexible workshop scheduling problems,this paper studies the flexible workshop scheduling problems from the perspective of single-objective and multi-objective respectively,proposes an improved whale optimization algorithm and an improved NSGA2 algorithm,and verifies the effectiveness of the algorithm through multiple sets of experiments,the core work of this article is as follows:(1)In this paper,a flexible workshop scheduling model with the maximum completion time as the goal is first established,in order to solve the situation that the whale optimization algorithm has low convergence accuracy and is easy to fall into the local optimal solution when solving the problem,a Whale optimization and Simulated Annealing(WOA+SA)fusion simulation annealing algorithm is designed.Firstly,the algorithm is tested on two sets of benchmark studies of Kacem and Brandimarte,and the test results are compared with ICSO,HGWO,Heuristic,IWPA and other algorithms,and the results show that the improved whale optimization algorithm has certain effectiveness in solving the single-objective flexible job shop scheduling problem with the maximum completion time as the goal.Then the algorithm is applied to solve two example problems,and the convergence accuracy of WOA+SA algorithm is significantly improved by comparing it with the traditional genetic algorithm,particle swarm algorithm,gray wolf algorithm and primitive whale optimization algorithm.(2)Aiming at the fact that multiple objective functions such as completion time,total machine load,and production cost need to be considered in the actual production workshop,a multi-objective FJSP model is established.To solve the problem of premature convergence or local convergence in multi-objective flexible job shop scheduling by traditional non-dominant sorting genetic algorithm(Non-dominated sorting genetic algorithm II,NSGA2),an improved NSGA2 algorithm is proposed.In the process of elite retention,the first N optimal individuals are replaced by the first N ×a individuals,and then the remaining individuals are randomly selected on the sub-optimal front surface.The algorithm caused by the decrease of population diversity is avoided to fall into local convergence.At the same time,neighborhood search is added to make up for the deficiency of traditional algorithm in local search.To prove that the proposed algorithm is feasible,this paper first establishes a scheduling model with the goal of completion time,total machine load,and maximum machine load.The proposed algorithm is tested on three standard test sets of Kacem,and the comparison of the test results with the artificial immunity algorithm,particle swarm plus taboo search algorithm and other four algorithms proves that the improved algorithm has a good effect in solving multi-objective scheduling problems.Finally,an example simulation proves that the improved NSGA2 algorithm has advantages over the original algorithm in solving the value of the objective function when solving this problem.The improved algorithm proposed in this paper mainly studies the static scheduling problem,and further research is needed on the dynamic scheduling problems such as order cancellation and delivery time advance in the production process,as well as the workshop scheduling problem based on digital twins.