Hybrid Predictive Maintenance for Building Systems Integrating Rule-Based Analytics for Proactive Energy Efficiency.
Hybrid Predictive Maintenance for Building Systems

Mazzetto, S. Hybrid Predictive Maintenance for Building Systems : Integrating Rule -Based and Machine Learning Models for Fault Detection Using a High-Resolution Danish Dataset.

A predictive maintenance framework based on machine learning techniques is proposed to provide guidelines to implement predictive maintenance for building installations and a case study showed the potential of this framework to predict failures.
Hybrid Predictive Maintenance for Building Systems

This study evaluates the effectiveness of six machine learning models, Artificial Neural Networks (ANN), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression (LR), for predictive maintenance in building systems . Utilizing a high-resolution dataset collected every five minutes from six office rooms at Aalborg ...
With the increasing focus on sustainable energy management and cost efficiency, predictive maintenance has gained traction as an essential strategy for modern building management [1,2,3]. Traditional maintenance practices, often based on fixed schedules or reactive responses, can lead to unnecessary downtime, inflated operational costs, and undetected system faults that contribute to energy ...

Hybrid Predictive Maintenance for Building Systems
Hybrid Predictive Maintenance for Building Systems : Integrating Rule -Based and Machine Learning Models for Fault Detection Using a High-Resolution Danish Dataset
Predictive maintenance , powered by advancements in machine learning (ML) and high-resolution data collection, offers a proactive approach, enabling building systems to operate more effectively by predicting faults before they impact functionality [6-8].

Overview of Hybrid Predictive Maintenance For Building Systems Integrating Rule
Mazzetto S. Hybrid Predictive Maintenance for Building Systems : Integrating Rule -Based and Machine Learning Models for Fault Detection Using a High-Resolution Danish Dataset.
The results indicate that combining expert-driven rules with machine learning and digital twin technology can deliver a practical, accurate, and scalable solution for predictive maintenance of AHUs. The framework supports the transition from reactive to intelligent building operations and can be adapted to similar non-residential contexts.
A hybrid deep learning approach to integrate predictive maintenance and ...
A rolling horizon planning strategy is adopted to continuously update the production and maintenance plans based on new data obtained through sensors. Unlike the existing integrated models, we develop a hybrid deep learning (DL) approach to coordinate maintenance and production decisions for a multi-state system composed of multiple machines.






This paper presents a hybrid approach to predictive maintenance tasks, which employs a combination of a supervised learning algorithm and a reinforcement learning algorithm.