Singapore Mobility Challenge 2022

Theme : More Efficient, Sustainable Operations

Early Detection of Bus Undercarriage Defects

Challenge Statement Owner: SBS Transit Bus

Data Analytics & AIAutomation & Robotics

Find out more from the Info Session!

Singapore has over 5,000 public buses – a significant part of our public transport network. Detecting defects in the buses early prevents incidents such as unplanned bus service downtime and accidents that could endanger commuter safety.

Currently, the inspection of the buses are done in workshops during non-operating hours. In order to inspect the bus undercarriage, which contains many key components, the bus needs to be hoisted using a skylift. The technician performs the inspection while referring to a checklist, and plans for further maintenance work based on the inspection. Critical defects that require longer servicing periods, assistance from other colleagues and/or availability of spare parts can cause delays to the bus returning to operation, especially if planning is not done in advance.

We are interested in solutions that automatically conduct the inspection of bus undercarriages during bus operating hours and meet the following requirements:

The inspection solution is designed to be conducted at designated locations that buses commonly pass and must be completed within a short period of time (within minutes).
The solution must take into consideration the bus undercarriage height from ground level of ~200mm and width between wheels of ~1,950mm.
After detecting the defects, they must be documented and channelled to the workshop team for maintenance planning. The documentation should include images of the defect. Critical defects must trigger an alert via email.
At the minimum, the solution should detect these critical defects: cracks on air bellows; wearing or looseness of the universal joints; and oil leakage. More information on other defects that could be detected are found under Resources.

How might we automate the inspection of bus undercarriage during operating hours, to aid the early detection of potential component failures and planning for maintenance work?

Resources