The SNCF reduces escalator downtime with AI-powered predictive maintenance
SNCF Gares & Connexions is a publicly owned limited company that specializes in railway stations—from design to marketing to operation. Part of the SNCF group, the company manages 15,000 train departures and 10 million passengers every day in 3,000 train stations across France. The company’s mission is to renovate, develop and modernize the SNCF’s vast network of stations, transforming them into convenient, efficient and welcoming places that are destinations of choice.
As part of its modernization project, SNCF Gares & Connexions paired with the City of Paris, Bouygues Energies & Services and Fieldbox to explore how predictive maintenance tools could be integrated into the operational maintenance of its escalators, in order to reduce downtime and improve the travel experience for passengers of the French railway network.
Escalators: A Moving Gateway to Travel
Every train station represents a gateway to the railway system, and where the SNCF travel experience begins. Escalators are an integral part of this experience, since most passengers are likely to take one or more of them before they board their train. Ensuring that all of a station’s escalators are operational and running smoothly is therefore key to providing a positive travel experience, keeping the flow of passengers continuously moving to avoid pedestrian traffic jams and overcrowding on the platforms.
Using Predictive Maintenance to Increase Uptime
SNCF Gares & Connexions wanted to see if they could improve the availability rate of their escalators by using automatic sensors and AI-powered predictive maintenance to prevent breakdowns before they happened. The idea was to reduce downtime by using AI technology to monitor components and detect the first signs of weakness that could lead to equipment failure, then replace those defective or failing parts en masse during quiet periods when escalators were less in demand.
The overall goal was to facilitate an unimpeded journey into and out of train stations, in accordance with the objectives set by transportation authorities, such as Ile de France Mobilités.
SNCF Gares & Connexions tasked French energy provider Bouygues Energies & Services with testing out the technical feasibility of using a predictive maintenance solution for its escalators, and chose FieldBox.ai to be its AI partner, based on the maturity of their technical solution and the company’s ability to work in tandem with business experts. The AI operator met all the requirements the SNCF was looking for: it was able to conduct an equipment diagnosis and determine what kind of sensors were needed for the job; build a monitoring device from scratch, to be mounted directly on the machinery; and possessed a proven track record in algorithmic science and machine learning.
The benefits of predictive maintenance
Get notifications when problems are likely to arise, so you can replace parts and schedule maintenance in time to prevent breakdowns.
Save Money on Replacement Parts
Optimize your purchase price by ordering replacement parts well ahead of time, instead of at the last minute.
Minimize unnecessary wear and tear on escalators related to poorly functioning machinery, so that they need fixing less frequently.
Optimize Equipment Efficiency
Schedule maintenance for low-demand periods when equipment is less needed, and extend the lifespan of escalators by using them more efficiently.
Destination: Gare de l’Est
The railway operations specialist chose Paris’ Gare de l’Est, one of the capital city’s six main railway stations, to conduct its experiment. The team conducted field visits, interviewing the station’s on-site technical experts and operational managers to understand their needs, how the equipment worked, the context in which it was operated, and the issues surrounding the relationship between the operator, maintenance service providers and operational staff in the SNCF stations.
Together with Bouygues and Fieldbox.ai, they installed a data capture device that could be accessed remotely and LORA Objenious sensors to measure temperature and vibration at different points along a selected escalator. Watteco NKE sensors were also used to measure the electrical intensity of the power supply from the escalator motor and a version of FieldBox.ai cloud software was deployed and configured for the project requirements.
Looking ahead, SNCF Gares & Connexions plans to continue experimenting with predictive maintenance tools, finding new and innovative ways to integrate them into their operational maintenance procedures for all their equipment. For SNCF Gares & Connexions, Bouygues Energies & Services and FieldBox.ai, the bigger goal is to use their findings to establish a predictive model that can be applied to equipment across the transportation industry and ultimately create a better and more positive travel experience for all.