This website can predict the destiny of your wait-listed train ticket

Indian Railways moves more people every day than the population of Australia.
Indian Railways moves more people every day than the population of Australia.
Image: AP Photo/Ajit Solanki
We may earn a commission from links on this page.

It’s one of the familiar dilemmas of life in India. What to do when you are wait-listed for a reserved seat in an Indian Railways train? Do you make other arrangements or just wait around hoping it will get confirmed?

Most people ask around and rely on what the extended family’s best-travelled uncle’s experience has been. Most frequent travellers have internalised an algorithm from experience—the likelihood of getting a confirmed ticket adjusted for several variables. A wait-listed ticket means you will move up in the queue and get a reservation when those who are confirmed for travel cancel or change their plans.

Now, a New Delhi-based software engineer is trying to take the guesswork out of this game.

Trainman is a three-month-old website created by 27-year-old engineer Mohammad Amir. Its algorithm can predict your likelihood of getting a confirmed ticket. Amir claims his predictions have an 85% accuracy record.

Wait-listed travellers have to enter their PNR number (passenger name record), which is generated after they book a train ticket on the Indian Railways website, and Trainman calculates their odds of getting confirmed passage.

Anything above 65% implies a high chance of confirmation. On the other hand, if the chances are below 50%, the website advises travellers to not hold their breath and try an alternative train or travel option. If users are looking for trains, Trainman helps them decide if they should book a waiting ticket or not, by predicting the confirmation chances.

Amir, a graduate from IIT Roorkee, has used a science called machine learning to build his website. It is a type of artificial intelligence that enables computers to act without being explicitly programmed. He collected historical data on various trains and fed the website’s prediction logic.

He says the website is getting about 10,000 users per month. A similar service,, claims a 73% total accuracy.