Railways switches to AI-enabled apps to address grievances & crowd management

Coach Mitra, meanwhile, is operational in 74 trains.

“We recently implemented an AI system in the RailOne app. It gives much more accurate information in the percentage on chances of getting a waiting ticket confirmed. Similarly, AI is already working in RailMadad, the grievance redressal system of Railways. It classifies and prioritises the complaints based on urgency and type,” said G V L Satya Kumar, Managing Director, CRIS on the sidelines of AI India Summit 2026 in Bharat Mandapam.

“We have integrated Bhashini to convert voice in 12 languages to text. It also identifies the complaint trends and gauges the passenger sentiment for proactive action,” he said.

Kumar said that CRIS is also working on AI solutions for predicting railway traffic congestion and accelerating the road to rail shift of freight using GST data.

“We are trying to build a system where Railways would be able to decide which new commodity to add to its freight basket. With the help of GST data, we have been able to map the movement of commodities from specific locations. On the basis of it, we have created a total of 300 freight clusters. These are the potential freight areas where Railways can move in,” said Kumar.

Another AI system in the pipeline is the development of an incident prediction model for the Railways assets using historical incidents and failure data. For the safe operation of trains, assets such as tracks, rails, locomotives, coaches, wagons and signalling systems have to function properly.

The AI system will predict in advance which assets need repair, moving away from the traditional maintenance practices that respond only after a failure.

CRIS is also working on the crowd management solution which will merge reserved ticketing data, unreserved ticketing data and train movement data to predict which specific platform of station will see higher crowds on hourly basis. The AI system will correlate ticket purchase timestamps with train schedules to visualise and map likely platform-specific footfall density.

It will also refine predictive models by accounting for festival logic, special events and specific weekend travel dynamics that deviate from standard patterns.

Officials said that the AI system will also identify the potential bottlenecks at station entry points and foot over bridges before they occur to ensure safe passenger flow.