Hong KongLogistics and transport2023 – ongoingMTR Corporation
Escalator Object Identification System
MTR installed escalator object detection with reported accuracy of over 95%
Problem
Small metal objects caught on escalator comb plates could jam equipment and create safety risks. Staff needed to find and remove them before they caused damage.
Approach
MTR's Capital Works Business Unit developed the system with Sebit. High-resolution cameras and a deep-learning model detected small metal objects on comb plates. The system sent mobile alerts so operations staff could remove the objects.
Results
Takeaway
Detection fed into a staff response workflow. MTR did not quantify the reduction in accidents or downtime.
Categories
- Size
- Large enterprise
- Industry
- Logistics and transport
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Computer vision
- Adoption
- Partner co-development
- Evidence
- Company's own claim
Sources
Source summaries
MTR Lab described an installed computer-vision system that alerted staff to metal objects on escalators. It reported accuracy of over 95% and qualitative reductions in maintenance costs and downtime.
Full details are in the original sources.
