Google Maps Experiment Shows Rerouting a Few Drivers Can Cut Congestion

The researchers collected weekday data and examined road types ranging from city streets to highways, according to the report. In Los Angeles and other places where the experiment targeted highways, diversions had the largest effect, increasing driving speeds in those corridors by about 4.6%. Across most of the other cities, speeds rose by only about 0.5% during peak periods.

The trial was designed to spread traffic more evenly across corridors when drivers headed to similar destinations. Scientists have long thought that pushing a few vehicles away from bottlenecks could ease congestion, and the results provided a real-world demonstration that the idea could work.

UCLA urban planning professor Michael Manville, who was not involved in the study, called it 'a pretty nice empirical illustration of a longstanding idea.' He said moving a small number of vehicles away from a clogged artery as it is about to enter a long delay can produce a 'not gigantic but noticeable result.'

The study's authors acknowledged potential limits. Alexandre Bayen, a co-author and director of UC Berkeley's Center for Information Technology Research in the Interest of Society, said this type of rerouting shares the risk of 'induced demand.' Just as adding a lane to a critical highway can temporarily relieve congestion until more driving reappears, a successful rerouting program could encourage more people to drive and erase the gains. 'That's why when we did the study, we only operated at the margin,' Bayen said.

The paper also estimated environmental benefits. Slightly smoother traffic and shorter travel times would cut carbon dioxide emissions by about 1,000 metric tons a year, equivalent to emissions from roughly 200 vehicles, a small share of the millions of cars in the cities studied. Bayen said the cumulative benefit would matter if such efforts were sustained over a longer period. Manville said the study served as a useful proof of concept: 'It is a demonstration that if we managed roads better, air would probably be cleaner.'

Bayen said city officials could decide to use the Google research findings for larger-scale congestion efforts, according to SFGate. Google declined the outlet's requests for comment on whether it had informed city authorities or rerouted drivers in connection with the study.