Using AI to prevent ticket gate breakdowns
Millions of people use ticket gates every day. But breakdowns cause disruption, frustration and lost revenue.
Traditionally, fixing ticket gates has meant waiting for something to break, then sending an engineer to diagnose the problem. But what if we could predict failures before they happen?
That’s exactly what Cubic Transportation Systems and Imperial College London’s AIDA Lab set out to do in this latest research via Imperial Consultants. The whitepaper shows how AI can be used to can spot issues early and keep gates running smoothly.
Predictive maintenance, proactive service
The team developed a predictive model that looks at error reports and forecasts which parts are likely to be needed after a failure is reported. This means engineers can arrive with the right parts in hand, ready to fix the problem straight away. In tests, the system predicted the correct part over 80% of the time and could cut engineer visits by up to 70%.
Keeping passengers moving
The benefits of an engineer arriving on site with the part needed to solve the issue are huge.
Rather than arriving, diagnosing the issue, ordering the part and then returning to fix the gate. An engineer could arrive with the right part and an understanding of the likely issue – reducing the number of call outs and cutting labour costs.
Quicker fixes also means less time with a broken gate. Resulting in better revenue collection, faster transit times through stations, fewer queues, and an overall better passenger experience.
Improving the reliability of fare collection systems not only helps train companies safeguard revenue but improves the reputation and, ultimately, usage of public transport.
What’s next for predictive maintenance?
This research is only the beginning. Our current model was trained on only a year of maintenance logs and focused on the 26 most frequently requested parts.
The long-term potential of this research goes beyond predicting parts needed for service from maintenance logs. This includes external factors, like bad weather or vandalism, that can impact gate reliability.
This research shows how data-driven solutions can make everyday travel better. By moving from reactive repairs to proactive maintenance, we’re building a transport network that works smarter for everyone.
To find out more, you can read the full whitepapers on Cubic’s website or visit the AIDA Lab at Imperial College London.