Multi-objective optimization method of product service systems configuration based on customer demand constraint mechanism

Abstract

This article proposes a multi-objective optimization method for product service configuration based on customer demand constraint mechanism, targeting the diverse customer needs of product service systems (PSS). This method first forms a utility matrix based on the response of product service instances to customer needs and establishes constraints based on product attributes and enterprise resources. The PSS configuration module is divided into product basic modules and optional modules based on customer demand constraint mechanisms, aiming to express the combination of product and service attributes. Through the utility matrix, multiple optional objectives such as safety, energy efficiency, and economy of PSS configuration are established to fully express the multi-dimensional needs of customer satisfaction. Then a new MOEA/D algorithm was developed for multi-objective optimization of PSS configuration. The improved MOEA/D algorithm adopts a weight vector adjustment strategy based on correlation sparsity, which improves the convergence and diversity of optimization results in high-dimensional objective space and solves the problem of exponential growth in solution space caused by the increase in customer demand types. This method was applied to practical cases of home appliance PSS configuration and compared with the performance of five advanced methods. The experimental results showed that the model established in this paper and the improved algorithm obtained configuration schemes had significant advantages in convergence and diversity.