HYDE PARK — Chicago is a vain metropolis, publishing every minute detail about the movement of its buses and every little skirmish in its neighborhoods. A team of researchers at the University of Chicago is taking that flood of data and using it to understand and improve the city.
“Right now we have more data than we’re able to make use of — that’s one of our motivations,” said Charlie Catlett, director of the new Urban Center for Computation and Data at the University of Chicago.
Over the past two years the city has unleashed a torrent of data about bus schedules, neighborhood crimes, 311 calls and other information. Residents have put it to use, but Catlett wants his team of computational experts to get a crack at it.
“Most of what is happening with public data now is interesting, but it’s people building apps to visualize the data,” said Catlett, a computer scientist at the university and Argonne National Laboratory.
Catlett and a collection of doctors, urban planners and social scientists want to analyze that data so to solve urban planning puzzles in some of Chicago’s most distressed neighborhoods and eliminate the old method of trial and error.
“Right now we look around and look for examples where something has worked or appeared to work,” said Keith Besserud, an architect at Skidmore, Owings and Merrill's Blackbox Studio and part of the new center. “We live in a city, so we think we understand it, but it’s really not seeing the forest for the trees, we really don’t understand it.”
Besserud said urban planners have theories but lack evidence to know for sure when greater density could improve a neighborhood, how increased access to public transportation could reduce unemployment and other fundamental questions.
“We’re going to try to break down some of the really tough problems we’ve never been able to solve,” Besserud said. “The issue in general is the field of urban design has been inadequately served by computational tools.”
Chicagoans already use the data the researchers are studying to make decisions about what bus to take using the CTA’s Bus Tracker or what route to drive home using traffic projections.
Catlett said people’s use of data about Chicago works well when its limited to these simple decisions, but it fails when we try to scale up.
“Intuition is based on experience and it can be really misleading if you try to apply it to something 20 times the scale,” Catlett said. “You can only go so far in your head because you can only hold so much in your head.”
Common sense does not reveal which workers on the South Side should start training now for new jobs in manufacturing.
“We would want to understand over time how you position people on the South Side of Chicago so that when light manufacturing comes in there are people with the training to work there,” Catlett said.
In the past, policy makers would make educated guesses. Catlett hopes the work of the center will better predict such needs using computer models, and the data is only now available to answer some fundamental questions about cities.
“Why is it that when we take two neighborhoods that appear to be similar over time one neighborhood becomes distressed while the other prospers?” Catlett asked, adding that answering such questions requires large and diverse datasets. “We haven’t had this amount of data until the recent years.”