Passenger information: a winning sprint
Twenty teams with many specialisations spent three weeks imagining the future of passenger information.
The technical incident at the Gare Montparnasse last summer and the serious traffic disruptions that followed were a reminder of the progress still to be made in passenger information at SNCF. Two months after the launch of the programme Rob.In (Robust Information), the company has gone a step further by organising Sprint Information Voyageurs,an event bringing together 54 internal teams and 62 European start-ups. The objective? To imagine new solutions for anticipating problems and better communicating with users. For the 20 finalist teams, the adventure ended on 13 November at Station F., the world’s largest campus for start-ups, located since June 2017 in the former Halle Freyssinet.
Developed in collaboration with Transilien, the Frankie project takes its inspiration for from the crowd-sourced navigation app Waze. Its ambition is to capitalise on the power of a community of 5 million daily commuters in the Paris region by enabling them to react to traffic conditions on their journey and report any unforeseen events (delays, incidents, exceptional numbers of passengers…), knowing that this information will be relayed in real time over SNCF’s various information systems and in particular the app. There will also be a space on Frankie for swapping alternative solutions. Passengers’ phones will thus be potential sources of mobility data. “The Innovation & Research teams have contributed the knowledge they have acquired in research projects related to the technological building blocks like AI and language processing that are essential to Frankie’s development. They were involved from the beginning of the project, when the Transilien division put forward the idea of a ‘Waze for transport’ at the Sprint Information Voyageurs”, explains Fabienne Réveillac, leader of the “Operating with artificial intelligence” project.
A multidisciplinary team of five experts in the scientific and technical network Synapses from Innovation & Research and SNCF Mobilités is working on the IV-PAAS (Passenger Information-Prediction As A Service) with the idea of putting data and learning techniques to use for passenger information. Combining historical and real-time data, its algorithms generate consistent and reliable predictions concerning delays, duration of incidents, and passenger numbers, which will be fed into the existing passenger information systems. They will open the way to new services enabling SNCF staff and passengers to act rather than simply react to the situations they are confronted with. Maguelonne Chandesris, the team leader, says: “IV-PAAS is based on several Innovation & Research prediction projects that are operational today in several areas. The idea is to make intelligent exploitation of data central to passenger information for the benefit of staff and passengers.” The Rob.In programme will contribute to these two projects.