Cubic Transportation Systems (CTS), an international transit software maker, is moving its real-time passenger information system to the cloud and working in machine learning capabilities.
Cubic’s NextBus product is essentially a platform for those agencies to show workers and customers how far away buses are from their next stop. The product has been around for 17 years and is used by more than 135 transit agencies in North America and Australia, but the next generation version Cubic is rolling out now is very different.
The new release more closely aligns NextBus with Cubic’s NextCity product, which supports single-account, digital payment for multiple means of travel. It also sets up ways for passengers to better plan multi-modal trips, where one might hop from a bus to a train to another bus.
And it is also embracing artificial intelligence to help bolster its prediction capabilities, which will feed into information passengers can access through multiple means. NextBus already had ways to guess at when the next bus might arrive at any given stop, but machine learning algorithms are increasingly used in the technology industry to make predictions like those more accurate.
That’s because machine learning algorithms are capable of ingesting massive amounts of data and finding relationships where humans might not think to look. In the future, NextBus might pull together bus locations, weather conditions, traffic jam reports and more to create a history-adjusted prediction of when a bus will get where it’s going. It might even be able to predict accidents, a concept gaining steam among transportation companies.
“When we train the predictor (algorithm), obviously we can allow it to train almost an infinite number of models to get the best fit, which a human could never do,” said Tony Gale, CTS’ general manager for NextBus.
The product also includes back-end analytics for transit agencies, using anonymized passenger data, to give them resources to answer questions. Gale described it as operational intelligence.
“It gives the ability of the transit operator to provide a much better service because they have a much higher visibility into their operation,” he said.
As it moves the product to the cloud, CTS plans on embracing application programming interfaces (APIs) and a modular approach to developing NextBus in the future. Though it offers some APIs now, Gale said he wants to build up a partner ecosystem feeding off the NextBus platform to power different services.
Gale said CTS is taking a phased approach to rolling out the overhauled NextBus, beginning with a few customers and then fixing any issues before offering it to others. Though he didn’t say what specific modules the company is looking to develop next for the product, that process should give CTS some insights into what customers want to see in it.
“We’ll learn as we go and perhaps make some refinements,” Gale said.
Ben Miller is the business beat staff writer for Government Technology. His reporting experience includes breaking news, business, community features and technical subjects. He holds a Bachelor’s degree in journalism from the Reynolds School of Journalism at the University of Nevada, Reno, and lives in Sacramento, Calif.