In 2010, the town of Mount Pleasant, S.C., implemented an adaptive traffic-control system in conjunction with ongoing improvements to U.S. Rte. 17.
The purpose of the system was to improve traffic flow and minimize driver delays both during and after construction. HDR provided the design plans and specifications for the installation and evaluated traffic conditions before installation. Further evaluations were conducted after installation both before and during U.S. 17 construction.
U.S. 17 is a major northeast-southwest highway running through Mount Pleasant, which is part of the Charleston metropolitan area. It is primarily a six-lane divided highway with left-turn lanes at all major intersections and right-turn lanes at some intersections and other access points. There are six signalized intersections included in the U.S. 17 study. Based on 2005 traffic data, approximately 35,500-43,000 vehicles travel this corridor each day. Recent peak-period traffic counts show volumes of 2,700-3,400 in the morning peak hour and 3,300-3,700 during the afternoon peak hour. The posted speed limit on U.S. 17 is 45 mph.
Retail developments are located along both sides of U.S. 17 between I-526 and S.C. 517, and there is a major shopping center along U.S. 17 and Hungryneck Boulevard between Market Center Boulevard and S.C. 517. In addition, Hungryneck Boulevard serves as a bypass for commuters, with destinations in the adjacent residential developments to the south and vacationers heading to the Isle of Palms from I-526. The traffic demand fluctuates between serving the retail and residential establishments located within the study area and seasonal vacation traffic heading toward the beach. This fluctuation in demand typically requires the development of multiple signal plans to address the varying traffic conditions.
The original intent of the study was to test operations of an adaptive signal system that would be deployed in conjunction with improvements to the roadway. The South Carolina Department of Transportation (SCDOT) wanted to use this information to investigate potential applications of adaptive traffic signals throughout the state.
Choice is Pleasant
As part of a roadway reconstruction project, Mount Pleasant was selected as a pilot project to test the state’s first adaptive traffic-signal deployment. The selection of this site provided the opportunity to test the system’s ability to address recurring and nonrecurring congestion within the town of Mount Pleasant. The InSync system by Rhythm Engineering of Lenexa, Kan., was installed in February 2011, prior to the widening of U.S. 17 through a section of this corridor. To the best of our knowledge, this is the first time an adaptive traffic-control system has been deployed intentionally for use during construction as well as after construction is complete.
SCDOT’s interest in adaptive traffic is shared by transportation departments across the country. Conventional traffic management requires signal operators using collected data to update signal-timing plans. Under the best of circumstances, the collected data provides only a narrow snapshot of traffic conditions, and often it becomes out of date soon after the plan is implemented.
The idea behind proper signal timing is to move traffic in platoons through a signal system. Adaptive traffic-control systems offer a more effective means of accomplishing that goal by continuously collecting traffic information with sensors, processing the data and adjusting signal-timing parameters to best serve the current conditions. In adaptive systems, the signal timings are typically updated every few minutes or few seconds for a certain few technologies. Compare that with conventional systems that operate set timing plans based on traffic-volume data that is several years old.
The system that was selected for the U.S. 17 study corridor uses digital video detection and does not require any in-pavement loop detection. It was installed at six signalized intersections on U.S. 17 and an intersection at Hungryneck Boulevard and Venning Road. The goal of the installation was to coordinate timing for the signals in a dynamic manner, taking into account real-time changes in traffic flows, such that vehicle progression on U.S. 17 was maximized and delays and stops were minimized.
Video-detection cameras were added to each signal approach (typically four per intersection). The system’s software uses the data collected by these cameras to call individual phases at specific signals and to adjust critical system operating parameters, such as cycle length, intersection offsets and green times. A processor was placed inside the traffic cabinet to communicate with the signal controller at each intersection.
As part of the study, an existing and future analysis was conducted using microsimulation in VISSIM to estimate the benefits of the adaptive system and identify any additional improvements required for the most effective deployment. Since being deployed, the system has operated very well. It is operating 24/7 and adjusting in real-time to traffic-flow changes along the corridor as construction gets under way.
The study measured travel time on U.S. 17 from Ira Road to Isle of Palms Connector, a distance of approximately 1.67 miles. Phase 1 travel data (before installation) was collected April 13-15, 2010, during peak traffic periods. Phase 2 (after installation, without construction) was conducted during peak periods on April 6, 7 and 12, 2011. Phase 3 (after installation, during construction) took place during peak periods on Feb. 8, 2012.
More than 140 travel runs were conducted. Due to weather conditions, an incident on the adjacent I-526 and the start of construction on U.S. 17 resulting in a small sample size for morning peak hour after installation, without construction, a comparison was not conducted during the morning period. Additionally, a midday period was added to the analysis for the after-installation runs both without and during construction.
Bump up the volume
The average travel times collected during the U.S. 17 study are shown in Table 1. With implementation of the adaptive signal system, the northbound travel time decreased slightly by 8 seconds (?4%) after installation during construction in the morning peak hour. The afternoon peak-hour northbound travel time decreased by more than 1 minute (?29%) after installation without construction and minimal increase during construction.
Average southbound travel time before implementation was estimated at 2:54 in the morning peak hour and 4:20 in the afternoon peak hour. After installation, without construction, the afternoon peak hour averaged a 43-second (?17%) decrease in travel time. After implementation, during construction, travel time increased slightly by 22 seconds (13%) in the morning peak hour and decreased by 53 seconds (?20%) in the afternoon peak hour.
As shown in Table 2, the average speeds on U.S. 17 prior to system implementation were between 24 and 35 mph during the peak periods. After implementation, without construction the average northbound afternoon speeds were 9.3 mph higher (35% increase). Southbound speeds after installation, without construction increased by 3.6 mph (14%). Speeds during construction generally decreased, with the exception of southbound afternoon peak-hour speeds, which increased from 25 mph to 30 mph (23%).
Before implementation, total average northbound morning stopped time was 38 seconds per vehicle and afternoon stopped time was 44 seconds per vehicle. Southbound average stopped time prior to implementation was 13 seconds in the morning and 1 minute 17 seconds in the afternoon (See Table 3). With the adaptive traffic system in place and without construction, average northbound afternoon stopped times improved substantially, with a drop of 32 seconds (?73%). Southbound afternoon stopped time decreased by 31 seconds (?40%). Results were mixed during construction, with some sample groups experiencing increased delays and others experiencing less stopped time.
Fuel consumption, vehicle efficiency and vehicle emissions
Post-installation data employed global positioning system (GPS) technology to facilitate more detailed time and distance measurements. The GPS data was processed to estimate fuel consumption, vehicle efficiency and vehicle emissions. As shown in Table 4, northbound and southbound vehicles experienced similar fuel consumption for the after-installation, without-construction peak periods, but northbound vehicles recorded substantially better fuel efficiency. Accordingly, the estimated northbound pollutant emissions are slightly less than for the southbound direction. This is likely related to the lower average stopped time in the northbound direction.
During construction, northbound and southbound vehicles again experienced similar fuel consumption; however, southbound vehicles recorded higher efficiency during the midday. As a result, the northbound pollutant emissions are higher than the southbound direction, with the exception of the afternoon peak where the pollutant emissions were higher in the southbound direction.
Vehicular counts were collected before and after the installation (without and during construction) and tabulated for each roadway segment along U.S. 17. One notable finding was that the average link volume for the southbound morning peak period during construction was slightly more than before installation, yet the average speed slightly decreased despite the presence of construction activities.
During the midday peak hour, the volume during construction increased along the entire corridor, and average speed increased by approximately 6%. Southbound afternoon volumes after installation increased slightly without construction and then decreased slightly during construction. Speeds increased both without and during construction.
The average morning northbound volume along the corridor slightly increased after installation and during construction while the average speed remained constant. Northbound volumes during construction generally decreased during the midday and afternoon peak hours. The midday and afternoon average northbound speeds after installation and without construction were significantly higher than speeds during construction.
Analysis of the U.S. 17 study suggests that implementing the adaptive traffic system yielded favorable results. The corridor experienced a number of travel-time and speed benefits after installation, and during the afternoon period in particular, average travel times decreased and average speeds increased in both directions. Unfortunately, the I-526 incident, weather and construction resulted in smaller-than-expected morning period samples. That said, the assessment of the system operation after installation, during construction suggested consistent and significant benefits in the reduction of travel time and increased average speed along the U.S. 17 corridor. In terms of volume and speed, the corridor experienced a slight decrease in average volume after installation, but generally showed improvements in average speed.
The project has met and exceeded the town of Mount Pleasant’s expectations. The pilot study proved to the town and SCDOT that the use of adaptive traffic-signal systems would provide a significant improvement over traditionally timed signals. The system has been successful enough to convince the town to continue to convert a significant number of existing signals along the U.S. 17 corridor, as well as along Hungryneck Boulevard, to adaptive-based traffic-signal systems.