Collaborative contracting for Manufacturing-as-a-Service (MaaS) by information content measurement and decision tree learning

• Published in: Advanced engineering informatics 2023-04, Vol.56, p.101911, Article 101911
• Main Authors: Gong, Xuejian, Wang, Shu, Jiao, Roger J., Gebraeel, Nagi Z.
• Format: Article
• Language: English
• Subjects: Crowdsourcing; Decision tree learning; Manufacturing-as-a-Service; Multi-criteria evaluation; Open design; Open manufacturing; Supply contracting
• ISSN: 1474-0346; 1873-5320
• DOI: 10.1016/j.aei.2023.101911

•Related to an emerging trend in Industry 4.0 towards product fulfillment through crowdsourcing.•Formulate contracting decisions in crowdsourcing product fulfillment as a best-matching problem.•An information content measure based decision tree learning of quantitative and qualitative metrics.•Dynamically configure a supply chain with appropriate design and manufacturing capabilities.•A case study of orthodontic brace mass customization through crowdsourcing. Product realization through Manufacturing-as-a-Service (MaaS) has observed as an emerging trend towards Industry 4.0. It offers new opportunities for reaching external partner’s knowledge and resources while allowing companies to focus on their core competencies. This paper envisions the importance of collaborative contracting for MaaS fulfillments, in which tournament-based crowdsourcing entails evaluation and selection of manufacturing service providers as a best-matching problem of multi-criteria decision making. MaaS collaborative contracting involves numerous evaluation criteria related to not only technical capabilities of the fulfilling agents, but also their business performance involved with MaaS operations in the past history. This paper develops a contracting analysis and evaluation methodology for selection of appropriate MaaS agent selection. The proposed evaluation mechanism utilizes information content measure and decision tree learning for better matching of fulfilling agents to maximize customer satisfaction and effectiveness of multiple stakeholders of MaaS operations. The multi-attribute utility theory is integrated with decision tree learning within a coherent predictive analytics framework to not only synthesize pre-defined functional and cost requirements for customers, but also leverage upon historical data of business performance. A case study of orthodontic brace mass customization through MaaS is reported to illustrate the feasibility and potential of collaborative contracting in MaaS.