Research On Project Scheduling Problem And Optimization Method Considering The Learning Effect Of Multi-skill Personnel

With the development of science and technology and market economy,under the fiercely competitive market environment,enterprises are facing increasing pressure and risks,and product research and development are increasingly valued by enterprises.For product R & D,companies often manage in a project-oriented manner,ordering R & D tasks through project scheduling and reasonably allocating resources in order to quickly and costeffectively launch high-quality products to win the market and increase corporate competitiveness.Due to the increasing number of product development tasks,the increasingly complex relationship between tasks and the presence of multi-skilled personnel,the number of feasible solutions to problems and the difficulty of solving them continue to increase.At the same time,the related skill learning of multi-skilled employees during task execution will lead to changes in skill levels to make it in a dynamic environment and the task duration is not fixed,problem-solving becomes more complicated,so project scheduling that considers multi-skill personnel learning becomes a difficult decision for enterprise manager.The project scheduling problem in software product development that considers the learning effect of multi-skill personnel is studied,analyzed and modeled and the problem-solving optimization algorithm is studied in this paper.Then,simulation calculation experiments are designed to verify the effectiveness of the algorithm and analyze the effects of the learning effect.Finally,the use of software product development examples illustrates the effectiveness of the research work and methods.Firstly,the project scheduling problem in product development considering the learning effect of multi-skill personnel is analyzed,and a model of personnel skill level change for the skill learning of personnel in product development is established.Then this paper formalizes the description of the project scheduling problem considering the multiskilled personnel learning effect and establishes a mathematical model.Secondly,the corresponding optimization methods for problem-solving are designed.To obtain a feasible and effective scheduling plan for the problem in a relatively short time and generate high-quality initial solutions of intelligent algorithms,this paper designs a heuristic algorithm based on priority rules.In the algorithm,the task selection and personnel assignment are based on the priority rules,and scheduling plans are generated under serial and parallel scheduling strategies respectively.In view of the lack of performance evaluation and effective use of search strategies in the problem-solving process of intelligent algorithms,to improve the efficiency of algorithm search,the Q-learning mechanism in reinforcement learning is used to guide the selection of algorithm search strategies,and the search strategies are designed in this paper.Finally,a dual-population co-evolution algorithm(QLCA)based on Q-learning is proposed.Then,simulation calculation experiments are conducted to analyze the performance of the heuristic algorithm based on priority rules under different combinations of priority rules and scheduling strategies,and the performance of the algorithm that can obtain higher quality solutions is verified.Then the latest two competitive algorithms and two classic algorithms are used as comparison algorithms to verify the performance of QLCA.The analysis of the experimental results shows that: QLCA is more efficient in search and obtains the Pareto solution set with better quality;at the same time,the comparative analysis of the personnel learning effect shows that the project scheduling considering the personnel learning effect has obtained a lower cost and a shorter duration.Finally,the feasibility of the research work and effectiveness of the algorithms are illustrated by an example of a bolt assembly management control project of a locomotive product.