IoT-Driven Digital Twin for Improved Product Disassembly in Remanufacturing

Remanufacturing, aimed at restoring end-of-life products to like-new condition, has become an essential value-recovery strategy to guarantee sustainable industrial development. Despite its innumerable social, economic and environmental associated benefits, the high uncertainty on the condition of end-of-life products represents a major barrier to the automation of disassembly tasks. Therefore, the effective adoption of remanufacturing in industries is limited these days. In this context, worldwide consolidation of the Internet of Things has led to the recent creation of massive data ecosystems, the potential of which is yet to be exploited in a remanufacturing domain. Data analytics and predictive planning built on top of the Digital Twin concept can significantly improve decision-making through the modelling of “what-if” scenarios leveraging real-time information collected and made available by Internet of Things network infrastructures. However, the remote monitoring of end-of-life products and the integration of information from multiple sources are not trivial tasks, which respectively deal with large-scale wireless network deployments and heterogeneous spatio-temporal patterns. To shed light on this matter, this work explores one of the recent advances of Digital Twin infrastructures towards remanufacturing, as well as open challenges and research directions in the field of data-driven decision-making. Second, relevant sources of information are discussed and reviewed, which are then classified into three hierarchical categories: (i) product-condition information, (ii) context information, and (iii) spatio-temporal features. Finally, a conceptual Digital Twin model leveraging multi-source information integration is proposed, which motivates a case study for data-driven decision-making in end-of-life product recovery considering context information. This work represents a step further in the adoption of the Internet of Things in the remanufacturing domain by addressing both data collection and exploitation dimensions towards Digital Twin modelling.