Tranquilien Combines Open Data and Crowdsourcing to Streamline Public Transport Flows

By July 04, 2013
public transport

Every day, three million people use the Transilien line and RER express commuter trains in the Paris region and half of them are equipped with a Smartphone. These passengers are now being encouraged to make use of a special app to feed information back from their train with a view to improving everyone’s general comfort.

The Tranquilien app is intended to give passengers a better chance of obtaining a seat on public transport. Developed by startup company :Snips in conjunction with French national railway company SNCF, the mobile app is designed to forecast passenger flows. This enables passengers to see how crowded the trains are and select, within the limits of their schedule, the train and even the specific carriage likely to provide the most comfortable journey.  This app may remind readers of the Dutch solution iNStApp, on which L'Atelier reported recently. Thanks to a similar, stripped-down, interface, “the passenger can follow a simple colour code: green – s/he should be able to find a seat; orange – some chance of obtaining a seat; and red – standing room only,” explains :Snips co-founder Rand Hindi.  However Tranquilien is not, like iNStApp, based on sensors tracking the number of people getting on and off the trains but on an algorithm which draws upon open data. It feeds through real-time passenger flow information for each line and also encourages passengers to play a useful role in traffic management by using the interface to assess the degree of comfort during their journeys.

App underpinned by open data and public collaboration

Rand Hindi points out that the app is a “pure product derived from open innovation and the SNCF open data initiative.” The algorithm being used is based on open data, supported by the passenger contributions plus the app’s use history. SNCF flow-data supplemented by demographic data on the various boroughs in the catchment area, plus SNCF’s historical usage information, enables traffic modelling and forecasts. The model is then adjusted to take account of passenger contributions coming in via the collaborative interface. “Crowdsourcing is an important element of the app. Users can tell us where our forecasts are wrong, thus enabling us to make corrections,” points out Rand Hindi.  As a final step, user history is fed into the algorithm, but he stresses that app data remains 100% anonymous. Interestingly, when the app users check in, “it’s not too serious if there’s no network coverage. The information is stored in the background and recovered as soon as the network is available.”

Exploiting data to improve daily life

Rand Hindi points out that it would certainly be feasible to extend the app’s use to other means of transport which has non-reserved seating – regional express trains, buses, etc. “However the question is: will the operators let us have access to their data? Still, for the moment the first step is to launch the app and promote the open innovation and open data approach.”  Currently, Hindi and his colleagues are “focusing in particular on public transport and urban mobility in general, because here we’re reaching saturation point and we need to find a solution.” Other uses one can envisage include tools to avoid having to queue up at the cinema or post office for example – in fact any kind of project where it might be useful to calculate the number of people doing something or other in a particular place at a particular time. “The basic aim is to optimise resources,” concludes Rand Hindi.


Legal mentions © L’Atelier BNP Paribas