DEVELOPING A THEORETICAL FRAMEWORK OF DIGITAL TWIN FOR INDUSTRIAL MACHINE FAILURE PREDICTION: A LITERATURE REVIEW

Abstract

Digital Twin has become a key approach in supporting predictive maintenance strategies for modern industrial machinery. Although numerous studies implement Digital Twin through numerical modeling and artificial intelligence, theoretical studies that map its conceptual structure, fundamental elements, and failure prediction mechanisms remain limited. This article presents a simple literature review aimed at identifying the core concepts of Digital Twin and developing a theoretical framework that can serve as a foundation for future research in mechanical engineering. Literature from Google Scholar, ScienceDirect, and IEEE Xplore was analyzed to formulate the structural components of a Digital Twin system within industrial machine contexts. The findings show that a Digital Twin is composed of a physical model, virtual model, data connectivity, and an analytics engine, all of which work integratively to detect anomalies and predict failures. The theoretical framework developed in this study is expected to serve as a reference for conceptual research related to machine maintenance and mechanical reliability.