As is often the way with project names, FAST is a play on words.
It is an acronym of Find A Space (on a) Train.
And the worked focused on using near real-time data to rapidly update of predictions of how busy a service will be further down the line. That is, what is the impact of the ‘fast’ collection and transmission of data.
The answer is: significant.
Real-time* data significantly improve the accuracy of crowding predictions, even when there are spikes in demand that the machine-learning models have not previously seen.
For passengers waiting on the platform, this means accurate insight into how busy they can expect the approaching service to be when they board and for the duration of their journey.
(*I should say ‘near-real-time’ – as engineers, we like to be precise about these things.)