A Knowledge Graph-Aided Concept–Knowledge Approach for Evolutionary Smart Product–Service System Development

TLDR

Conventional PSS development relies on many professional consultations and fails to support Smart PSS features such as user context‑awareness and evolving knowledge management. The paper proposes a cost‑effective evolutionary design approach for Smart PSS by integrating knowledge graph and concept–knowledge modeling to meet user expectations and enhance flexibility and sustainability. Two knowledge graphs built from open‑source knowledge, prototype specifications, and user text are used to trigger four KG‑aided C‑K operators that, via graph‑based queries and NLP, generate innovative solutions tailored to personalized requirements. A case study on a smart nursing bed demonstrated acceptable knowledge‑evolution performance, suggesting the approach provides useful guidance for industrial Smart PSS development.

Abstract

Abstract In order to meet user expectations and to optimize user experience with a higher degree of flexibility and sustainability, the Smart product–service system (Smart PSS), as a novel value proposition paradigm considering both online and offline smartness, was proposed. However, conventional manners for developing PSS require many professional consultations and still cannot meet with the new features of Smart PSS, such as user context-awareness and ever-evolving knowledge management. Therefore, aiming to assist Smart PSS development cost-effectively, this paper adopted the knowledge graph (KG) technique and concept–knowledge (C-K) model to propose an evolutionary design approach. Two knowledge graphs are firstly established with open-source knowledge, prototype specifications, and user-generated textual data. Then, triggered by personalized requirements, four KG-aided C-K operators are conducted based on graph-based query patterns and computational linguistics algorithms, thus generating innovative solutions for evolving Smart PSS. To validate the performance of the proposed approach, a case study of a smart nursing bed fulfilling multiple personalized requirements is conducted, and the evaluation result of its knowledge evolution is acceptable. It hopes that this work can offer insightful guidance to industrial organizations in their development of Smart PSS.