A hypergraph-based approach for context-aware smart product-service system configuration. (January 2022)

• Record Type: Journal Article
• Authors: Wang, Zuoxu; Chen, Chun-Hsien; Zheng, Pai; Li, Xinyu; Song, Wenyan
• Is Part Of: Computers & industrial engineering. Volume 163(2022)
• Publication Date: 2022-01
• Subjects: Smart product-service systems -- Configuration system -- Hypergraph -- Context-aware -- User-centric design; Engineering -- Data processing -- Periodicals; Industrial engineering -- Periodicals; 620.00285
• DOI: 10.1016/j.cie.2021.107816
• Languages: English
• ISSNs: 0360-8352

Abstract

Highlights: Introduced a hypergraph-based approach to deal with the complex many-to-many relationships. Proposed a requirement attribute-product-service bundle-usage scenario data model. Provided an unbiased hypergraph ranking algorithm to avoid unbalanced weightings on hyperedges. Abstract: Smart product-service system (Smart PSS) is a heterogeneous and integrated system in which manufacturers/service providers deliver integrated and customized product-service bundles (PSBs) in a sustainable manner. Smart PSS configuration, a process of selecting proper PSBs based on user requirements, is expected to be further flexible to adjust the configuration results based on users' straightforward requirements that contain user-preferred usage scenarios and technical attributes. However, the technical attributes provided by users might be inaccurate due to a lack of professional knowledge, thus prone to unaligned configuration results. Besides, sorely technical attributes other than a comprehensive scope of information was emphasized in the conventional PSS configurators. To address these problems, this paper proposes a novel hypergraph-based Smart PSS configuration framework. Contrary to the conventional configurators that emphasize the mapping between technical attributes and PSBs, the proposed framework introduces usage scenario information as the auxiliary information, which is usually straightforwardly expressed by users. To represent the heterogeneous information with less local information loss, a hypergraph model based on the requirement attribute-product-service bundles-usage scenario (RA-PSB-US) data model is established. Besides, a hypergraph ranking algorithm is adapted to rank the PSBs. To mitigate the proneness of selecting settled PSBs other than returning adaptable results (i.e., the bias of hypergraph-based Smart PSS configuration) during the configuration process, an unbiased hypergraph ranking algorithm is proposed by normalizing the hyperedge degree. Finally, a comparative study is conducted based on an online 3D printing service review dataset to validate its effectiveness and advantages.