Configuration optimization of service solution for smart product service system under hybrid uncertain environments. (April 2022)

• Record Type: Journal Article
• Authors: Chen, Zhihua; Zhou, Tongtong; Ming, Xinguo; Zhang, Xianyu; Miao, Rui
• Is Part Of: Advanced engineering informatics. Volume 52(2022)
• Publication Date: 2022-04
• Subjects: Smart product service configuration -- Multi-objective optimization -- Rough-fuzzy data envelopment analysis -- NSGA-II; Computer-aided engineering -- Periodicals; Engineering -- Data processing -- Periodicals; 620.00285
• DOI: 10.1016/j.aei.2022.101632
• Languages: English
• ISSNs: 1474-0346

Abstract

Abstract: Configuration of service solution is recognized as a crucial task for smart product service system (PSS) development, which provides an effective way to meet dynamic customer requirements. New smart attributes, broader stakeholder value and hybrid uncertainty occur inherently in the service solution configuration for smart PSS. These emerging characteristics lead to urgent updating need of the existing approaches to service solution configuration optimization. Therefore, the current study proposes a systematic approach to optimization of smart product service (SPS) solution configuration, aiming to achieve the best smart capability and value symbiosis among stakeholders of the configured SPS schemes. In the proposed approach, a new attributes system is proposed to describe the smart service module instances, and then employed as criteria to determine the configuration parameters. A novel rough-fuzzy data envelopment analysis (DEA) method is proposed to obtain the configuration parameters of each service module, with objective of effectively handling the hybrid uncertainties involved in the configuration environments. In addition, a bi-objective optimization model for SPS solution configuration considering smart capability efficiency and value symbiosis efficiency is proposed, and it is solved by an adaptive non-dominated sorting genetic algorithm II (ANSGA-II) to acquire a Pareto set of optimized configuration solutions. Finally, an application of this systematic approach to smart vehicle service demonstrates the feasibility and effectiveness of the proposed approach.