Big Data vs. Traffic Jams
All rights reserved: Les Echos/Worldcrunch
Looking for a parking spot is not only frustrating, it’s also one of the causes of urban traffic gridlock. In fact, as much as a fifth of all traffic stems from drivers trying to park their cars, author Donald Shoup suggests in his book The High Cost of Free Parking.
With mobility now a crucial concern in most big cities, it’s no wonder that many startups are turning their attention to the problem. To take up the challenge, the San Francisco-based company Smarking relies on Big Data.
The startup provides private parking lot owners with a clear and understandable overview of their data. “Every time a driver pays, the gate goes up to let him in. Later, when he wants to leave, it’ll go up to let him out. That’s how we generate data,” Cassius Jones, growth manager for Smarking, explained during the Demo Day at this year’s Bridge SF event.
It is thus possible to know how full a parking lot is, depending on the time of day and of year, to measure how long a car stays on average, etc. “We store this data in the cloud and we present it to our clients in a holistic and intelligible way, in clear and explicit diagrams, so that they can make the right decisions,” Cassius Jones said.
Higher prices during the high season
Smarking works in partnership with the city of Aspen, Colorado, where tourists flood in during the winter and summer periods, bringing the traffic downtown to saturation point. By increasing prices in parking lots in overcrowded areas during the high season, the city was able to cut parking occupancy rates by 15% and to decongest the city center. In return, the city of Aspen has lowered the prices during quieter seasons.
In New Haven, Connecticut — also one of Smarking’s partners — authorities realized that some parking lots were always fuller than others. To optimize usage, they therefore increased prices in the saturated ones and lowered them in the more deserted lots.
Other companies, like Cloud Park, which also took part in the Demo Day, combine machine learning and video cameras to make available parking spots visible, thus allowing municipalities to optimize their parking pricing.