Billions of train journeys are fuelling Trainline's latest travel tool an analytics engine that aims to introduce transparency to train companies' practice of hiking ticket prices up.
The Price Prediction feature launched yesterday in the Trainline app and aims to save customers up to half of the cost of advance tickets by analysing a multitude of past journeys.
This allows it to find patterns of demand and determine when prices will rise, refining its approach as app users book tickets and make new journeys.
Jon Moore, chief product officer at Trainline, said: "Our data scientists have used historical pricing trends from billions of customer journey searches to predict when the price of an Advance ticket will expire.
"We now share this information in our app to allow our customers to get the best price possible for their journey. We're introducing more advanced machine learning every day so naturally our predictions will get increasingly accurate. Our mission is to make train travel as simple as possible and price prediction is the first in a long line of predictive features we have planned to help customers save time and money."
Trainline said the tool could cut 49% off the price of advance tickets, compared to 43% before they built the tool.
One example it cited was a standard class advance single fare ticket from London Euston to Manchester Piccadilly. Costing 32 if booked 80 days before the day of travel, Trainline found the price rises to 38 41 days before the day of travel, increasing up to 126 if booked on the actual the day of travel.
Using Price Prediction, the app can then tell users that they can save up to 88 if they book the ticket 80 days before they travel.
Price Prediction is the latest machine learning-based feature Trainline has launched. Another recent innovation, BusyBot, crowdsources customer information about where they're sitting to help people find seats on busy trains.
The company is making more and more use of AI, running limited trials of a tool on its website that predicts where customers are travelling based on their ticket histories, the time of day they're making the booking, and when they want to travel.
CTO Mark Holt previously told IT Pro that Trainline is working on pulling in contextual data to help people book the most convenient journeys.
He explained: "If we know Manchester United are playing at home that Saturday you're trying to book a ticket with your four children to go to Manchester, we might suggest you go after 3pm when it will be a little less full with football supporters. That sort of thing is so great from a customer experience perspective - it's a real value add."
Pictures courtesy of Trainline
