Research On Methods For Low-carbon Flexible Job Shop Scheduling Considering Worker Learning

Flexible job shop scheduling problem(FJSP)is a typical Non-deterministic Polynomial hard(NP-hard)problem,which aims at the effective management and allocation of workshop production resources,so as to improve the manufacturing efficiency and market core competitiveness of enterprises.Although the research on FJSP in recent years has received extensive attention from academia and industry,due to the complexity and changeover of the actual processing situation in the workshop and the difficulty of solving the problem itself,many constraints that may exist in actual production need to be further studied.Firstly,the worker learning effect is one of the constraining factors which widely exists in the actual production and has an important impact on the production and manufacturing.Secondly,the proposition of the concept of green manufacturing makes most enterprises start to consider the manufacturing factors of green environmental protection while paying attention to economic benefits.How to balance the economic benefits and environmental impacts of enterprises is an urgent problem for many enterprises.In addition,how to design an efficient optimization algorithm to solve the newly proposed LFJSP-WL model is also one of the key and difficult points in this research.Therefore,it is of great academic and practical significance to study a production scheduling model and optimization algorithm that balances worker learning and low carbon emission.In view of the above problems,the main research contents of this article are as follows:(1)A low carbon FJSP model considering worker learning is proposed(LFJSPWL),in which the workers have the ability of autonomous learning,which has a certain impact on the workpiece processing time.According to the difference of learning ability and cumulative processing time of each worker,the processing time function of each worker is established.In addition,the carbon emission caused by workpiece processing,machine standby,tool wear consumption and auxiliary material consumption is further analyzed,and the corresponding function model is established.By considering the above two functions,a mathematical model is established to minimize the maximum completion time,the total cost of workers and reduce the total carbon emission of the production process.(2)A memetic algorithm is designed to solve the proposed LFJSP-WL model.According to the characteristics of the problem,three effective initialization methods are designed for each of the three sub-problems of the model to improve the quality and diversity of the initial population,and two crossover operators and three mutation operators are proposed according to the structure of the chromosome,which aim to optimize the quality of the solution and the space of the solution during the evolution process.In addition,a variable neighborhood search operator with four neighborhood structures is designed to enhance the local development capability of the algorithm,the elite pool is used to update and retain the outstanding individuals generated during each evolutionary process,and the optimal combination of key parameters of the algorithm is obtained by means of Taguchi orthogonal experiment.By extending the benchmarks of the FJSP,22 verification examples suitable for the model in this paper are constructed,and the performance of the proposed algorithm for solving the LFJSPWL model in this paper is verified by multi-group comparison experiments.(3)A web prototype system for workshop scheduling which is easy to expand is developed.Through the analysis and summary of the mathematical models and optimization algorithms of related workshop scheduling problems,a set of workshop scheduling prototype systems suitable for different workshops and multiple scenarios is developed based on Javaweb technology.The system includes four modules: system management,workshop data management,scheduling management and report management.