Nothing is more frustrating for a commuter than to sit in traffic on the ride to work or home. The same goes for the bus rider who must contend with too few buses that are packed during rush hour. Congestion is one of the evils of modern living, causing stress and irritation. It’s also hugely expensive, costing the country $100 billion annually in lost time and productivity, according to Professor Nancy Folbre of the University of Massachusetts.
Stockholm, London and Singapore have resorted to congestion pricing as a way to reduce traffic by forcing drivers to pay for the privilege of driving downtown during the busiest work hours. Other cities, mostly in South America but also in China, have resorted to road rationing: limiting cars on the road according to odd and even license plate numbers during the week.
But as Balaji Prabhakar pointed out, both of these solutions act as a stick, penalizing drivers, which makes them hugely unpopular. Just ask former New York City Mayor Michael Bloomberg, who tried to introduce congestion pricing to midtown Manhattan when he was in office, only to scuttle the plan after strong opposition.
Prabhakar, who is a professor at Stanford University and director of the Stanford Center for Societal Networks, believes there’s a better way. He and Shiva Shivakumar, a former executive with Google, have founded a company called Urban Engines that emphasizes spatial analysis and behavioral economics to reduce congestion on roads and in transit systems. Urban Engines works with city transit authorities to figure out better ways to use existing infrastructure and to craft incentives – carrots rather than sticks – to change people’s commuting habits and reduce congestion. “We attack urban congestion with a combination of insights about the underlying traffic model and incentives to shift behavior,” said Shivakumar.
The company uses spatial analysis to create a digital replica of a city’s transportation system and helps cities implement incentives based on behavioral economics that reward commuters for shifting their travel away from peak times. “Paying people to not travel at peak hours has proven very effective,” said Prabhakar. In a pilot project conducted in Banglore India, where traffic congestion has reached epic levels, incentives involving lottery tickets convinced a respectable 17 percent of commuters to shift their travel time away from peak hours.
So far, Urban Engines has worked with transit agencies in three cities – Singapore, Sao Paolo and Washington, D.C. – using data from smart cards to create a data model for spatial analysis. The smart cards also provide a mechanism for incentives. Train passengers could be offered discounts on their cards if they used them during nonpeak time periods, according to Prabhakar.
In America, where the vast majority of commuters drive by themselves in cars, the sources for data and incentives are different. The nation’s major tolling systems provide traffic data that can be analyzed and become the basis for behavioral incentives, such as discounts credited to accounts, for driving at less congested times.
Another strategy for reducing congestion, according to Prabhakar, is to look for hotspots. “It’s useful to know how, when and where the congestion hotspots occur,” he said. “That leads to a better understanding of how supply is used and making the supply more flexible.” For example, a bus route may have a hotspot that affects just 20 percent of the route’s length during rush hour. To reduce wait time and overcrowding, transit agencies can send out more buses that run along just the most congested portion of the route. “If you make supply more responsive to demand or incentivize the demand, you can change commuting behavior,” he said.
The challenge to reducing congestion is the constant, near daily, anomalies that can make predicting traffic and congestion patterns difficult. Prabhakar calls them “nuances.” From bad weather to a sporting event at a stadium, nuances can throw normal traffic patterns askew, making matters worse.
To address the issue, Urban Engines founders Shivakumar and Prabhakar say their system reconstructs what a large transit system does from the edge, using a wide range of data sources, including images from cameras, to show how the entire system is working and then applying behavioral economics to shift driving and transit riding behavior. “It’s about making supply more responsive to demand,” said Shivakumr.
This piece was originally published at Government Technolgy magazine.