Condition-based and Predictive Maintenance Strategy for Lift Installations using Big Data Analytics

Safe and reliable lift services are essential for maintaining high accessibility and functional vertical transportation to help preserve the vitality of cities like Hong Kong which are renowned for its densely packed skyscrapers. This paper presents a proof-of-concept trial of a health monitoring platform for condition-based and predictive maintenance of lift installations using big data analytics. Implemented with various non-intrusive sensors, time series data of temperature, strain, acceleration, and displacement of lifts are collected and used to build predictive models with statistical and machine-learning techniques. The novel approach is capable of fault detection of brake malfunctions, lift car shaking, door malfunctions, and traction motor malfunctions and potentially enables prediction of the remaining useful life for the critical components.