Research On Optimal Allocation Model And Algorithm Of Mobile Human Resources In Multiple R & D Projects

In recent years,China has transformed from high-speed development to highquality development.The country pays more and more attention to the investment and development of R&D projects.The development of R&D enterprises and users’ demand for product iteration speed greatly increases the number of R&D projects.And the parallel operation of multiple projects has become a new feature of R&D enterprises.As the main resource of R & D projects,R & D personnel often need to participate in the R & D work of multiple projects in the enterprise at the same time and transfer and switch between tasks of projects due to their limited number.From a practical point of view,the R & D personnel transfer between projects need to spend some time getting familiar with and preparing for the new project background and progress,which is called transfer time in this paper.Different from the traditional physical transfer of machinery and equipment between projects,the transfer of personnel between R & D projects has no relationship with the geographical location,so it is called non-physical transfer.Personnel transfer may lead to the delay of the task and affect the project duration and scheduling scheme.Therefore,managers in the enterprise need to pay attention to it.At the same time,there are differences in the technical level and project experience of R & D personnel,which is called personnel heterogeneity.Personnel heterogeneity affects task duration and transfer time.It is embodied in two aspects: first,people with the same skill have different skill levels,so their work efficiency is different,and the time required to complete the same work is also different;Second,there are differences in the experience accumulation of personnel on related projects,so the time to get familiar with the same project or task is different,that is,the transfer time is different.The impact of R & D personnel heterogeneity makes the staffing problem of multiple R & D projects more complex.Therefore,the research on the optimal allocation of mobile human resources in multiple R & D projects has both theoretical significance and practical value.In the actual project management process,in order to effectively shorten the R &D period and improve the economic benefits of enterprises,enterprises prefer to obtain better project schedule and human resource allocation scheme through intelligent scheduling optimization methods.In this context,considering the impact of nonphysical personnel transfer and heterogeneity on project scheduling,this paper proposes an optimal allocation model of mobile human resources in multiple R & D projects and design the corresponding algorithms to expand the research in the corresponding field.Firstly,for the optimal allocation model of mobile human resources in multiple R& D projects,a detailed problem is described,a mixed integer programming model is established.And the feasibility of the model is verified by CPLEX with simulation data.Secondly,according to the characteristics of the model,a parallel schedule generation scheme-human transfer time(PSGS-HMTT)which considers personnel transfer time is proposed;Based on PSGS-HMTT,twelve heuristic algorithms are designed as candidate methods by combining 3 task priority rules and 4 transfer priority rules.The best heuristic rule combination min LFT+max Skill is selected through simulation instances of different scales.On the basis of PSGS-HMTT,4 meta heuristic methods based on genetic algorithm are designed as candidate methods by combining two initialization methods and two coding methods.Simulation instances of different scales are used to verify the solution effects of the 4 meta heuristic methods,and the best meta heuristic algorithm HRIGA-TRL is selected,which is better than the best heuristic algorithm in the previous step.Two adaptive mutation improvement factors are designed to improve the performance of meta heuristic algorithm.The improvement factor is applied to HRIGA-TRL to form HRIIMGA-TRL for simulation experiments to verify the effect of the improved meta heuristic algorithm which is further improved than the original meta heuristic algorithm.The final results of simulation experiments show that,compared with other heuristics and meta heuristics,HRIIMGA-TRL significantly shortens the total duration of multiple projects in all scale simulation examples,and has a potential to be applied and popularized in practical production.Finally,heuristic and meta heuristic algorithms are applied to solve the instance formed by the data obtained from the survey of three parallel R & D projects in NX software service company.And HRIIMGA-TRL which can get the shortest duration is verified that have the best effect in solving the real instance compared with other algorithms.The project scheduling scheme and human resource allocation scheme of the case is proposed according to the output results of the algorithm.