The Train Brain is a forecasting service that predicts delays hours before they happen. By monitoring real time arrival and departure times, the forecast model knows when a train is late and calculates the risk of further delays across an entire rail network. 



For traffic control purposes onboard computers and check points throughout a railway system transmit data to traffic control centers, identifying the positions of each train. This data is run through the algorithm The Train Brain which creates a continuum of forecasts, that can be accessed through an API.

The forecasting model can be compared to an earthquake seismograph; it looks for significant peaks. Trains pulling into a station behind schedule represent a peak. The algorithm automatically compares how a similar delay on the same line impacted the network on previous occasions, to predict how it will affect the network in the near future. And like a human brain it evolves. It learns and improves its ability to make forecasts over time.



In order to really help commuters the forecasts need to be updated very quickly. A train network is affected, every minute,  by a multitude of events. Some are scheduled and some just happen randomly, like accidents or someone illegally running across the tracks. This means that even if a forecast is almost 100 % accurate when delivered, it may be obsolete only 10 minutes later. One of the main features of The Train Brain is its speed. It recalculates and continuously adjusts its forecast in less than 1 minute.


How do forecasts help COMMUTERS? 

The Train Brain forecasts were designed to give commuters a heads up when their daily commute is likely to be disrupted. The key is to get this information at a time when you can act on it. Not at the station. Not on the bus heading to the train station. But well before you leave your home or workplace, when there still is time to arrange for an alternative mode of transportation. 


How early can accurate forecasts be produced?

Accurate forecasts can be produced as early as the time it takes to travel the longest service line in the rail network. If the longest line takes 2 hours to travel, the forecasts can be delivered up to 2 hours before scheduled departure. 


How do forecasts help traffic control?

The Train Brain forecasts can be used as an input source for traffic planning. For example as input to a traffic simulator, where traffic controllers can evaluate scenarios bases on projected consequences of a manual intervention. It is also possible to integrate the forecasts with other data sources such as current weight of car sections. This would provide traffic control with a decision tool that visualises the number of passengers in each train, making it possible to make decisions based on the amount of travellers affected by an disruption of service.



The Stockholm installation runs on version 1 of the forecasting model. Ongoing tests and development indicate that the next version of The Train Brain will also be able to forecast bus delays.