A Study of Critical Passenger Throughput Rate of Escalators and their Safety in Metro Stations

Project: Research

View graph of relations


Over 4 million people a day ride on around 1000 escalators in Hong Kong’s metro system.With approximately 16.8 million escalator rides, around four people persons are injuredeach day giving a total of approximately 1,400 injuries per year. These accidents resultedin serious delays in metro operation and extremely stressful and heavy burden on themetro operators and engineers causing multi-millions of losses financially to the metrocompany in particular and the society at large. Due the lack of standardized approach ofprediction of accidents, currently, metro engineers could only modified the hardware ofescalator systems as much as they could to improve their safety. Safety officers can onlypost various safety messages, in the form of posters and voice announcements, to raisepassengers' safety awareness when using the escalators inside metro stations. However,accidents on escalators remain top of the metro system's accident list. We recentlyconducted a study in which the passenger throughput rate (PTR) of a station was foundto be a major contributing factor to the escalator-related accident rate (EAR). Normally,the EAR is low when the PTR of the station is low. The accident records of a sample ofmetro stations in Hong Kong were investigated. It was found that a critical PTR exists atwhich the EAR reaches its maximum. Further increases in the PTR reduce the EAR. Itwas also found that this critical PTR is not a constant but varies for different stations.Therefore, in addition to the PTR, other station factors must contribute to thedetermination of the critical PTR. The research team aims to identify these factors foreach station and build a model to show the correlation between the PTR and the EAR.Because these factors may be related to the design of the station and other humanrelated activities, and no prior knowledge of the correlation is available, an intelligentapproach is proposed to mimic the correlation between them. Following the completion ofthis project, a real-time intelligent-based monitoring model will be developed to predictthe critical PTR at different times for the existing stations or the design of new stations.When the condition of a station becomes unfavourable and approaches the critical PTR,the prediction model will alert the station officers to take appropriate action. Thisintelligent approach will reduce the accident rate on metro station escalators and henceassisting to alleviate financial and societal losses.?


Project number9042227
Grant typeGRF
Effective start/end date1/01/1622/06/20

    Research areas

  • Artificial neural network,Escalator safety,Metro station,Transportation,