Multi-service value chains collaboration for repairperson resources selection using a many-objective evolutionary algorithm with adaptive reference vectors

Abstract

When facing with rising demand for repair services and technical problems, service stations with limited resources suffer a decline in service quality. However, high-quality service is essential for customer retention. The purpose of this paper is to solve the problem by sharing and allocating repairperson resources across multi-service value chains in order to improve the quality of after-sales repair services. Firstly, based on the multi-service value chain collaboration model in a third-party cloud platform environment, a repairperson resources selection model is proposed to balance the multi-service value chain repairperson resources, considering the interests of resource users, resource providers, and the third-party cloud platform simultaneously. Secondly, a many-objective evolutionary algorithm with adaptive reference vectors(EAARV) is designed to solve the resources selection model with an irregular Pareto front. Finally, a case study is conducted to compare the performance of EAARV with seven state-of-the-art evolutionary optimization algorithms for solving many-objective optimization problems and to validate its viability. The experimental results show that EAARV outperforms others in solving the repairperson resources selection problem, and the selection model considering multi-service value chains collaboration is proven to promote the utilization of service resources. The satisfaction rate of repairperson resources with multi-service value chains collaboration is significantly higher than the single-service value chain collaboration. Meanwhile, the parameters of EAARV are analyzed and an ablation experiment is conducted to further evaluate the influence of each component in EAARV on the performance.