Digital Twin-Driven Predictive Maintenance for Autonomous Manufacturing Systems

Abstract

This study aims to develop and evaluate a Digital Twin-driven predictive maintenance framework to enhance reliability and operational efficiency in autonomous manufacturing systems. A qualitative approach was employed using a case study design, selected for its ability to capture complex interactions between cyber-physical systems, predictive analytics, and digital twin technologies within a real-world industrial context. The research was conducted in an Industry 4.0-based manufacturing facility in East Java, Indonesia, chosen due to its partial implementation of IoT-enabled automation and digital monitoring systems. Data were collected from twelve participants, consisting of eight key respondents and four expert informants, selected purposively based on their technical expertise and involvement in maintenance and digital transformation processes. The findings indicate that integrating digital twin technology with predictive maintenance significantly improves failure prediction accuracy, real-time monitoring, and autonomous decision-making. The framework also addresses critical challenges such as data fragmentation and limited system interoperability. These results demonstrate that the proposed approach effectively reduces downtime and enhances system adaptability. The study recommends the adoption of standardized data architectures and the integration of advanced analytics to support scalable implementation in broader industrial environments.