Toronto aims to use data for traffic insight

Share

Although the sense that traffic is a problem was potent enough that John Tory tapped it to help win the Toronto mayoralty last fall, the city had little idea – and no current data – about what was really happening on key roads. Until now.

Toronto's Transportation Services department has enlisted an academic to analyze travel data from GPS providers for parts of 2011 through 2014, wading through the mass of information for useful nuggets.

This wealth of new information is part of the city's push into using big data. And it forms the basis for a 48-hour "hackathon," beginning Friday evening, aimed at finding ways to use this data to improve the situation on the roads.

"This kind of data, really it's a new insight into the transportation system," said Matthias Sweet, the assistant professor in Ryerson University's school of urban and regional planning who crunched the numbers for the city. "Getting your finger on the pulse of some of the unique characteristics of the transportation system is a way to move forward and evaluate it."

Although the data in hand don't yet extend into this year, and some of the results are obvious, the historical information offers the most detailed look yet into the city's traffic situation.

"With big data, you have good insight into what's actually happening in the streets," said Edwin Kools, a traffic analyst for TomTom, one of the providers of the data Toronto is using, who has come from Amsterdam for the event.

"I would say that Toronto is ahead of the pack and the fact that they are organizing this hackathon proves that."

This weekend's event at the Evergreen Brick Works is expected to attract 150 people in the city's first large-scale attempt to use the data. Stephen Buckley, general manager of the Transportation Services department, said the city hopes to build relationships with the tech industry, look for ways to tap "this new world of information" and get the value of a fresh set of eyes.

"It will be interesting to … cut that many people loose just to see what they can come up with," he said.

Because the material does not include data from 2015, it can't offer insight into whether a towing blitz championed by Mr. Tory had any impact on travel times.

Among other things, the data offer long-term trends on speeds on city highways and arterial roads. And the findings show that the picture is more complicated than the general public narrative about traffic being uniformly awful and getting consistently worse.

Although some of the city's highways do show declines in peak period average speeds from 2011 to 2014, others show improvements. For example, while the evening speed on the eastbound Gardiner Expressway dropped 16 kilometres per hour, the morning speed rose by 7 km/h. And the peak period speeds on the westbound Gardiner rose by 12 km/h and 15 km/h in the morning and evening peaks, respectively.

In general, though, average speeds on the city's expressways are down. Speeds on the arterial roads are also down, by about the same amount as the highways, though the percentage drop is greater because they started from a lower speed.

Another interesting finding is that while the fluctuation of highways speeds indicates they still have recognizable morning and evening peak periods, arterial roads behave differently. On the arterials, speeds drop through the morning and then stay down right through the day, indicating that heavy use is no longer concentrated in the traditional rush hours.

Mr. Buckley said that access to the kind of detailed data now available can have far-reaching effects on how the roads are managed.

"We can start quantifying," he said. "You know, we can say, if we have to close a facility down, what are the impacts? What are the best days, what are the worst days?"