-
-
Demo 1
-
Demo 2
-
Demo 3
ORM want to help rail passengers better understand the impact of disruption to their journey. How does disruption manifest? How does it affect other services/operators? Is there a butterfly effect to a disruption event? Are some services/stations more prone to disruption? Can other data help understand disruption? What is the likelihood of delay to passengers?
Introducting Traccine. We treat delay as a disease... And Traccine is the Vaccine!
From our analysis of Darwin data, we know that:
- If an incident occurs, a station can get ‘sick’ which can infect other stations on the network and lead to delays.
- These incidents need to be ‘contained’, ‘controlled’ and ‘rectified’ ... And that’s what Traccine enables
In treating delay as an epidemiological challenge rather than a mathematical one - we account for the structure of the rail network. As such the AI we’ve implemented is simpler and quicker, but just as effective as other more complicated approaches
In this way we can predict ‘outbreaks’ of delay earlier and respond quicker
People are becoming more data literate:
- We also aim to democratize our data to offer their customers greater insight into their journey
- We can embed our early warning system into existing apps and systems offer real-time predictions around delay severity & resolution resulting in a better UX for the customer
We’ve also developed a platform for TOCs to manage their delay also by building an AI that can:
- Predict delay + shockwave it sends through the network
- Can also predict recovery
The challenge we were set was around transport users, however we feel that the real prize here is helping rail and transport companies At the moment, it’s an early warning system but as we gather more data we may also be able to predict the best way to manage incidents and suggest solutions (in conjunction with TOCs). This can help rail companies and Network rail monitor, manage & deliver their services There's huge opportunity to apply to wider transport - roads, aviation flight routes, shipping lanes…
Log in or sign up for Devpost to join the conversation.