A machine learning-based iterative design approach to automate user satisfaction degree prediction in smart product-service system

As an emerging digital servitization paradigm, smart product-service system (Smart PSS) leverages smart, connected products and their generated services to work as a solution bundle to improve individual user satisfaction. As a complex solution bundle at both system and product level, its iterative design differs from the existing ones mainly in two aspects. Firstly, massive in-context data during the usage stage can be leveraged to calculate the satisfaction degree of individual users intelligently. Secondly, Smart PSS, consisting of both digitalized service and physical components, can be changed in a more flexible way in a data-driven manner. An iterative design method for fast positioning and replacing the unsatisfied modules can improve the user experience and extend the Smart PSS usage life. Nevertheless, some studies made attempts, and it is still missing an iterative design method with automatic real-time user satisfaction prediction. Aiming to fill this gap, this work proposes a machine learning-based iterative design approach to automate user satisfaction prediction in the Smart PSS environment. Furthermore, an illustrative case study of a surgical robot for flexible ureteroscopy is demonstrated along with this proposed methodological framework, which overcomes the challenges of subjectivity and tedious assessment of the experts in the conventional approaches. This research can offer some valuable guidelines to today's industrial companies in Smart PSS development.