University of Minnesota Driven to Discover
Universities MUniversities Wordmark
Center for Transportation Studies Heading

Webinar Available

Visualization and Assessment of Arterial Progression Quality Using High-Resolution Signal Event Data and Probe Vehicle Travel-Time Data

photo of Darcy Bullock

Darcy Bullock

Darcy Bullock, Purdue University Department of Civil Engineering

December 2, 2010

Coordination is essential to providing the highest possible quality of service on signalized arterials through movements, and travel time is its most accurate yet most expensive measure. In this seminar, Professor Darcy Bullock presented a methodology for evaluating signal coordination that combines high-resolution signal controller data with travel-time measurement using Bluetooth-device Media Access Control (MAC) address matching. He also described the Purdue Coordination Diagram (PCD), a tool for visualizing large amounts of controller and detector event data, quantitatively evaluating signal performance, and identifying opportunities for improvements.

The PCD, Bullock explained, plots the arrival time of each vehicle at an intersection using input from setback detectors combined with phase state (red and green intervals) information. On a cycle-by-cycle basis, it is possible to view the arrival of each platoon relative to the start and end of the green interval. At a higher level, the performance of the green band can be qualitatively evaluated by visually inspecting concurrence, or lack thereof, of vehicle platoons within the green bands. Quantitative measures such as the percent of vehicles arriving on green can be extracted from aggregation of the data. The impact of offset adjustments can be predicted by manipulating PCDs for an arterial corridor.

Bullock described how these techniques were demonstrated in a case study involving a before and after comparison of an offset tuning project. A signal offset is a signal timing parameter that substantially affects arterial travel times; it refers to the time difference between the start of the signal’s green interval and a system reference time. The offset is considered “bad” when vehicles arrive at a signal light just as it turns red; an optimized offset enables cars to arrive when the light is green.

In this study, arterial travel times were measured using MAC-address matching via intersection and mid-block detecting stations to identify time periods in need of timing adjustments. Optimal offsets were calculated for a Saturday plan on a four-intersection signalized corridor, and the operational impacts were estimated. These offsets were then implemented. PCDs were used to identify causes of poor progression in the before case, as well as to visualize both the predicted and actual arrival patterns associated with the optimized offsets. More than 300 Bluetooth probe travel-time measurements were used to statistically assess the before and after travel time. “Instead of modeling the probe data, we’re actually measuring [them]. In the past we relied on modeling because technology constraints limited the data we could collect,” Bullock said. All of this is changing as technology capabilities improve by leaps and bounds and costs decrease, he added.

The assessment [of before and after travel time] is really a statistical distribution, he continued. “Some vehicles get through on green, some don’t, some just barely make it through…We can look at this as a histogram of the travel time down the corridor and compare the statistical distributions using a cumulative frequency distribution (CFD).” In this case, the statistical comparison showed a significant one-minute reduction (approximately 20 percent) in mean travel time for the northbound direction, and a significant 0.5-minute (approximately 10 percent) reduction in mean southbound travel time. “We achieved a reduction in travel time just by optimizing four offsets, with no impact on capacity,” Bullock explained, adding that the number of vehicles in the corridor arriving on green increased from 56 percent to 66 percent.

Bullock’s team was also able to illustrate the economic impact of the reduced travel time. It found that the offset changes resulted in fuel cost and the value of time savings at the four intersections of $150,000 per year for a 1.6-mile stretch of roadway. In an age of dwindling financial resources, he said, it’s a welcome opportunity to show DOT executives the bottom-line benefit that investing in signal-timing improvements provides.