Application of Dynamic and AI Approaches for Predictive Maintenance

The paper deals with the application of AI for predictive maintenance. It is the result of multidisciplinary ongoing research conducted by the Department of Architecture (DiARC) of the University of Naples Federico II, Department of Civil Engineering and Architecture (DiCAR) of the University of Catania, and ETT S.P.A., a leading company in Digital Transformation. The objective of the research is to demonstrate how new Information and Communication Technologies (ICT) technologies, in the era of Smart cities, can be used not only for managing and monitoring new buildings but also in the context of built heritage to protect it, through an optimized approach for the planned maintenance. ICT can revolutionize maintenance, especially from Facility Management perspective. In this sense, Artificial Intelligence (AI) and Internet of Things (IoT) systems can optimize maintenance processes, enabling knowledge automation. It is possible to delineate statistical models related to built heritage performance over time by analyzing, interpreting, and systemizing Big Data from heterogeneous sources, moving from reactive to proactive approaches. The method involves a first phase characterized by a condition-based maintenance strategy and then moves to a predictive maintenance strategy. By learning from itself through the use of AI, the proposed system predicts degradations and failures even before they occur: the asset preservation guarantee is increased, and consequently, so is the useful life. Testing AI on a case study will also demonstrate how much the application of automation systems in predictive maintenance processes can lead to savings in economics, time, material, and labor compared to traditional scheduled maintenance systems.