Is it worth developing Digital Twins for Industrial Maintenance? Current Trends and first Assessment Elements

Digital Twin (DT) has emerged as a cornerstone technology in industrial maintenance, providing dynamic virtual representations of physical assets that enable real-time monitoring, predictive analytics, and performance optimization. While its technical benefits are widely acknowledged, questions persist regarding its actual profitability and environmental sustainability. Based on previous works, this paper provides an updated overview examining how emerging technologies, particularly artificial intelligence (AI) and augmented reality (AR), are reshaping DT applications in industrial maintenance contexts. It synthesizes the latest academic and industrial advancements to outline key trends, methodological innovations, and representative case studies across multiple sectors. A specific focus is placed on the convergence of machine learning and AI-driven predictive maintenance with immersive visualization technologies, which together improve data interpretation, operational decision-making, and operator training. In this study, several dimensions are considered: the maturity level of DT implementations, the types of industrial equipment observed, the quantity and nature of data collected (sensor data, operational parameters, environmental variables), and the specific maintenance strategies applied (corrective or preventive, the latter encompassing timebased, condition-based, and predictive approaches). The analysis also compares applications across sectors such as energy, manufacturing, and transportation to evaluate their respective levels of adoption and return on investment. Beyond conceptual and technical dimensions, the study seeks to determine the conditions under which Digital Twins constitute sustainable and cost-effective solutions for industrial maintenance. To address this issue, an evaluation framework is proposed, integrating both qualitative and quantitative metrics, including economic and environmental costs, to assess performance throughout the DT life cycle, from design to operation.
This dual perspective enables a holistic understanding of how DT generate value across different industrial ecosystems. The results of this research are expected to provide strategic insights for industrial stakeholders, thereby supporting evidence-based decision-making regarding technology investment, operational optimization, and the transition toward more resilient and sustainable industrial systems.

Keywords: Digital twin, industry, maintenance, augmented reality, virtual reality.