Design of Elevator Fault Monitoring and Early Warning System Based on Internet of Things

Elevator systems are critical in building operations, yet fault detection and predictive maintenance remain challenging. Internet of Things (IoT) technologies can enhance monitoring by determining the present information and predictive knowledge in enhancing safety along with operational efficiency. This research aims to design an elevator fault monitoring and early warning system using IoT sensors and a novel Rhizostoma Optimization Fine-Tuned Scalable Support Vector Machine (RO-SSVM) for accurate fault detection, prediction, and real-time alerts. The dataset, sourced from Kaggle, includes timestamp data from IoT sensors monitoring elevator door systems. It comprises brake status, humidity, and vibration sensor data, along with fault categories (no fault, minor fault, and critical fault), providing valuable information for fault detection and predictive maintenance. Data preprocessing using min–max normalization is employed to remove noise and handle missing values, ensuring the quality and consistency of the sensor data. Discrete Wavelet Transform (DWT) is applied to the time-series sensor data to extract features in both time and frequency domains. The RO-enhanced SSVM hyper-parameter demonstrates superior performance in identifying elevator system faults, ensuring reliable and efficient predictive maintenance. The system is implemented in Python, with the RO-SSVM model achieving high performance, with accuracy (99.82%) and AUC (99.30%), demonstrating effective fault detection and early warning capabilities. The proposed IoT-based elevator fault monitoring system demonstrates a significant improvement over traditional methods, offering early warning capabilities and predictive maintenance insights, which can reduce downtime and enhance elevator safety and reliability in smart buildings.