Investigating determinants of digital twins for predictive maintenance

Purpose In Industry 4.0, different technologies are used to improve the efficiency and reduce the downtime of the processes in the organization. It can be achieved by using predictive maintenance (PdM) technique to avoid the sudden breakdowns in the industry. It is important to implement digital twin (DT) for PdM. DT for PdM is in the nascent stage. This study is focused on the identification of determinants of the DT for PdM for real-life implementation. Design/methodology/approach This study has focused on the determinants of DTs for predictive maintenance for implementation. To analyse these determinants, multi-criteria decision-making (MCDM) techniques were used by applying Decision-Making Trail and Evaluation Laboratory (DEMATEL) and interpretive structural modelling (ISM) approaches. Findings In this study, 13 determinants were found out through the literature survey. These determinants were classified into cause and effect in the DEMATEL approach. Similarly, ISM methodology was applied to these determinants to categorized determinants into different levels. DEMATEL and ISM results were compared, and it was found that real-time analysis, decision-making, self-monitoring and diagnosis are the most important. Practical implications This study is useful for the academic researcher as well as the industrialist to implement DT in PdM. Therefore, it can be implemented in real-life application by considering the important determinants. Originality/value This is one of the first studies that represent the investigation of DT for PdM using the DEMATEL and ISM approaches.