The goal was to determine just how successful Siemens SCOOT adaptive signal control (ASC) system has been for this city of 115,000 residents, which swells to 185,000 when the University of Michigan is in session. Results of the first test were significant: SCOOT decreased average weekday travel times by 12% and weekend travel times by 21%.
By 2015, Ann Arbor already had a decade of experience with SCOOT (short for Split Cycle Offset Optimization Technique). Before SCOOT was deployed in 2005, it was difficult to keep traffic moving in the city. On an average weekday, 130,000 commuters arrive in Ann Arbor for work, while another 40,000 leave it to work elsewhere. Special events happen any day or night of the week, drawing thousands. Football games at Michigan Stadium can double the number of cars in town. SCOOT eases congestion on all of these occasions by using sensors and a computer algorithm to dynamically change traffic signals based on real-time conditions.
These manual studies could cost between $100,000 and $200,000 each and covered a very limited time period. In 2015, the Center of Excellence worked with StreetLight Data, a mobility analytics provider that transforms trillions of geospatial data points from GPS and cellular devices into actionable metrics. SCOOT performance studies could now be conducted at a much lower cost and cover longer time periods.
The Ann Arbor study took place along the Ellsworth Corridor, a two-mile stretch that was not part of the original SCOOT implementation. SCOOT was added to the corridor in November 2015, a month after the road had been re-timed. A single-lane in each direction, the corridor experiences high traffic volumes from people avoiding congestion on the nearby Interstate 94. It was the perfect place for a before-and-after study. The results showed that SCOOT reduced weekday travel times from 236 seconds before its implementation to 207 seconds afterward. Weekend travel times decreased from 232 seconds before to 183 seconds after implementation. SCOOT also significantly increased the likelihood that drivers would meet their target travel times.