Abstract
A project to install an Active Traffic Management (ATM) system on Virginia Interstate 66 (I-66) from US 29 in Centreville to the Capital Beltway (I-495) began early 2013 and was completed in September 2015. This project installed smart infrastructure and employed dynamic operations techniques that mobilized ATM, which were intended to improve safety and operations on I-66 without physically expanding the roadway. The main components of ATM that were installed on I-66 include advisory Variable Speed Limits (VSL), Queue Warning Systems (QWS), Lane Use Control signs (LUCS) and Hard Shoulder Running (HSR). ATM has been successful in producing safety and operational improvements in many European countries, but there are limited ATM applications in the U.S. (Mirshahi et al., 2007; Fontaine and Miller, 2012). Since ATM is still a relatively new approach in the U.S., there was a need to analyze the safety and operational effects of ATM on I-66.
In this thesis, appropriate operational and safety measures of effectiveness (MOE) were defined, examined, and analyzed in order to conduct a before-and-after study to quantify the effectiveness of the ATM system on I-66. The operational MOEs included ATM utilization rate, average travel time, travel time reliability, and total travel time delay levels. The safety MOEs included crash rates by type and severity and a safety surrogate (speed drop events) for crashes. These MOEs were analyzed by using INRIX travel time data, limited traffic volume point sensor data and Virginia Police Crash Reports.
The results indicate that the ATM produced positive operational and safety benefits in several MOEs analyzed in this report. The analyzed operational and safety benefits from implementing ATM on I-66 were similar to the reported operational and safety benefits of ATM implementations in Europe and other states in the United States. The research found that ATM generally had a limited operational impacts during the weekday peak periods on I-66, but some benefits were observed during the off peak weekday periods. Average weekday travel times during the middle of the day and in the off-peak direction typically improved by 2 to 6 percent. Large benefits were observed on the weekends, with average travel times improving by about 10 percent during the day. All of these differences were statistically significant. Travel time reliability improved by similar margins. Weekday peak period travel times and reliability continued to degrade following ATM installation, however. This was not surprising given that HSR was already in use during the peak weekday periods and there has been a historic trend towards increasing travel times on the corridor. Estimates of safety impact based on limited empirical data and safety surrogate analysis showed a 10 percent improvement in crashes during the weekdays and up to a 50 percent improvement on weekends. Those safety findings are preliminary, however.
High-level segment analysis was performed to determine the segments that benefitted the most from the implementation of the ATM. From this analysis, it was found that the HSR was the component of the ATM that produced the most operations and safety improvements on I-66. In terms of HSR utilization rate, HSR was being opened for a longer period of time after the implementation of ATM. On weekdays, the shoulders were open for an extra 2.5 hours/day and on weekends, the shoulders were open for an extra 4.5 hours/day for both directions combined. Before HSR, the shoulder opening hours were static during weekdays limited to peak periods and were not open during weekends.
A planning level benefit-cost ratio was calculated based on the operational and safety benefits. The ATM project had a benefit-cost ratio of 5.29, and its value was calculated by monetizing weekend operations and safety improvements on I-66. The high benefit-cost ratio shows that the ATM was a cost-efficient solution in improving operations and safety on the I-66 corridor. The thesis concludes with recommendations for additional expansion of ATM in Virginia and future research.
