U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
This magazine is an archived publication and may contain dated technical, contact, and link information.
|Publication Number: FHWA-HRT-13-003 Date: March/April 2013|
Publication Number: FHWA-HRT-13-003
Issue No: Vol. 76 No. 5
Date: March/April 2013
Researchers are working on microsimulation modeling to improve mobility and reduce congestion.
|Shown here is a typical work zone on I–495 in Virginia. A temporary concrete barrier along the leftmost lane shifts from a 10-foot (6.7-meter) offset, commonly used to shield workers from nearby traffic during construction, to the edge of the left lane.|
As the Nation’s highways continue to carry increasing numbers of travelers, State and local departments of transportation (DOTs) face the growing challenge of minimizing the impacts of work zones on traffic congestion. According to the 2011 Urban Mobility Report, congestion-related delay nationwide has grown from about 4 billion hours in the year 2000 to 4.8 billion hours in 2010, with the cost of wasted fuel and lost time reaching $101 billion.
Part of the solution involves improving the management and operation of highways. DOTs have a number of strategies at their disposal to reduce delay: incident management, ramp metering, coordination of signals on arterial streets, access management on arterials, and use of high-occupancy vehicle lanes. Together, these treatments resulted in an estimated savings of 327 million hours in travel time in 2010, compared to 190 million hours in 2000, according to the 2011 report.
The Federal Highway Administration (FHWA) estimates that nearly half of nonrecurring congestion is caused by temporary disruptions such as incidents, work zones, or inclement weather. Work zones in particular cause about 24 percent of nonrecurring congestion and 10 percent of all congestion. As such, work zones affect the overall mobility of a highway, having a reach that extends well beyond the limits of the construction area.
To test and evaluate operational strategies that might help reduce congestion in work zones, engineers use analytical tools such as modeling and simulation to forecast traffic conditions. However, says Jack Klodzinski, supervisor of travel forecasting at the Florida Department of Transportation’s Turnpike Enterprise Headquarters, more research is needed to help engineers understand the best ways to “dynamically address congestion issues as the construction progresses.” With the necessary data, he says, researchers could identify and test potential solutions using traffic simulation models.
Toward that end, researchers with the Saxton Transportation Operations Laboratory at FHWA’s Turner-Fairbank Highway Research Center in McLean, VA, and at the University of Central Florida, Orlando, are studying driver behavior as a means to generate data to improve the modeling and simulation of traffic flow at freeway work zones. Their research examines differences in driver behavior to better calibrate models and improve the accuracy of estimates of work zone impacts on traffic flow. Ultimately, this work aims to help analysts and practitioners choose and use appropriate calibration parameters during modeling and simulation of alternative work zone designs, as well as operational strategies to reduce impacts on mobility and safety.
According to Klodzinski, this research is an important step toward mitigating congestion around work zones. “Driver behavior is very understudied,” he says. “It is a parameter of simulation modeling that needs further research. I experience nonrecurring congestion every day that could potentially be improved if the data for use in a simulation model were available.”
Simulating Behavior In Work Zones
Traffic engineers use a variety of tools to better understand the effects that work zones have on traffic, and selecting the right one to perform the analysis is important. When a low volume of input data is available, engineers can use the Highway Capacity Manual, QuickZone, Quewz software, and impact analysis spreadsheets developed by individual States. For more complex projects and more detailed analyses, engineers turn to microsimulation, which involves the modeling of individual vehicle movements on a second or subsecond basis to assess traffic flow.
To model the systemwide traffic impacts of work zones, engineers can use microsimulation software programs such as AIMSUN, CORSIM, Paramics, and VISSIM. Using these models has drawbacks, though, such as the considerable time and expertise they require, as well as the large amount of data input necessary to construct a model. However, once a microsimulation model is assembled, it can provide engineers with an accurate virtual environment for testing new strategies. Evaluating these strategies can help engineers develop better traffic management plans for work zones.
Calibrating such detailed models is a challenge for researchers because the changes due to driver behavior in work zones are not fully understood. Greater insight into the behavior of drivers and vehicle dynamics in work zones, and incorporating this knowledge into the models, could vastly improve their validity. Similarly, the availability of model parameters specific to work zones could enable engineers to make more accurate predictions of the effects of operational strategies in a work zone environment.
Currently, work zone capacity is not a direct input in simulation models. Engineers typically adjust parameters such as car-following, lane-changing, and speed-choice algorithms to replicate traffic flow in work zones. For the FHWA study, Calibration of Driver Behavior Parameters for Optimization of Work Zone Throughput, researchers designed an experiment with the goal of capturing data on changes in driver behavior that occur in work zones. The experiment required the use of an instrumented test vehicle to collect data on driver behavior needed for calibrating psycho-physical, car-following models.
Instrumenting the Vehicle
To collect and analyze data on driver behavior, researchers equipped a sport utility vehicle with sensors to capture information on the gaps between vehicles and the speed oscillations of drivers. When following another vehicle, the space a driver leaves between his or her vehicle and the vehicle in front is defined as the gap. The variability of this gap over time is defined as the oscillation. The gap oscillations are a result of the accelerating and braking movements that a driver makes when following a vehicle.
|To collect data on driver behavior in traffic, researchers equipped this sport utility vehicle with a variety of sensors, including radar equipment being installed behind the front and rear bumpers.|
The onboard equipment used in this experiment includes a GPS-based speed sensor, multiple cameras with a video acquisition system, two universal medium-range radars, and a computer, which stores the GPS and radar data. The radar equipment, mounted behind a plastic bumper to avoid detection by other drivers, works in conjunction with the speed sensor to collect the relative distance from and velocity of the vehicles in front of and behind the instrumented test vehicle. The video acquisition system logs activity gathered from the forward-facing and rear-facing video cameras located inside the vehicle. The video footage helps the researchers analyze the radar data and identify what is going on around the instrumented vehicle. For example, researchers can watch the video to see when other vehicles change lanes, which is difficult to discern by simply looking at the radar data.
|In this screen capture of the drive recorder software used in the research, the surrounding vehicles are shown in the target view window on the right and also can be seen in a video window at the top center of the screen. The software logs the data collected by the instrumented vehicle at a rate of 40 points per second from the radar and displays the information in rows at the lower left of the screen.|
Drive recorder software logs data from the front and rear radars and the GPS-based speed sensor. The drive recorder is an off-the-shelf technology modified to fit the needs of this particular research. The software logs the objects that the radar identifies into several classifications by number, and the objects are shown as different colors in the software interface. The drive recorder can track the vehicles in front of and behind the instrumented vehicle. Other objects that the radars identify, such as roadside objects, are logged for future analysis but are shown as circles in the target view window in the software interface. The program logs distance, speed, accelerations, and decelerations at a rate of 40 data points per second. The researchers used the instruments to record data on the behavior of various test drivers as they passed through a work zone.
Conducting the Experiment
FHWA recruited participants to drive the instrumented test vehicle through an actual, real-world work zone -- a living laboratory -- set up along a 3-mile (4.8-kilometer) stretch of I–95 between Springfield and Lorton, VA, in coordination with FHWA’s Virginia Division and the Virginia Department of Transportation. A living laboratory is defined as a roadway, corridor, or network that is instrumented for measuring the performance of transportation system operations. This study is one of several that researchers at the Saxton Transportation Operations Laboratory are conducting in freeway living laboratories in coordination with State and local DOTs and university researchers.
|Dana Duke (left), with contracting firm SAIC, and FHWA Research Engineer Taylor Lochrane set up the living laboratory along a section of I–95 between Springfield and Lorton, VA, to capture traffic data.|
Participants were not told the exact purpose of the study; in fact, the study was described to the participants without mentioning the phrase “work zone” because of concerns that it might bias their behavior. Each participant learned about the actual purpose of the experiment upon completion of the trip.
Beginning in February 2013, 64 volunteers drove the experimental vehicle through the study area and afterward completed a short questionnaire documenting their comfort levels while driving through the work zone. The experiment focused on collecting data from the drivers for the entire duration of their trips, through portions of the roadway segment both within and outside the work zone areas.
It took an average of 2–3 hours to collect data from each participant. Researchers gave participants about 30 minutes to familiarize themselves with the vehicle prior to beginning the trip. The data gathered from the followup questionnaire provided researchers with demographic information on the participants, along with their responses to questions about how they felt driving through the work zones.
Online Survey Results
The researchers also released the study’s questionnaire online through a variety of social media outlets, asking for volunteers to complete the survey anonymously. The goal was to obtain opinions from drivers from across the United States regarding their comfort level driving in work zones.
Of the 738 people who fully completed the survey, 69 percent reported feeling “very comfortable” driving on a freeway in general, 26 percent reported feeling “somewhat comfortable,” and the remaining 5 percent described themselves as “not that comfortable.” Among the “very comfortable” drivers, there was a negative shift of 50 percent to “somewhat comfortable” and a shift of 17 percent to “not that comfortable” when entering a work zone. These findings indicate that motorists who are typically comfortable driving on the freeway are less comfortable when passing through work zones.
Focusing on these “very comfortable” drivers, the data reveal that the presence of cones, barrels, and temporary concrete barriers all show a similar reduction in comfort levels, with around 45 percent of these drivers reporting that their comfort levels drop to “somewhat comfortable” when traveling in a work zone. However, for cones and barrels, the data reveal a 14 percent reduction to “not that comfortable” and, for temporary concrete barriers, there is a 32 percent reduction to “not that comfortable.” These findings show that the reduction in comfort level is due to common roadside equipment, especially temporary concrete barriers. This supports the assumption that drivers may change their normal driving behavior while traveling through a freeway work zone.
The researchers hypothesized that car following would be the main influence on throughput and capacity in modeling results and therefore did not focus on lane changing. This assumption was supported by the questionnaire results, with most drivers (57 percent) reporting that they remain in the same lane while in a construction zone and maintain the speed of the vehicle in front of them. Maintaining the speed of and distance from the car in front is the unconscious car-following behavior that the instrumented vehicle sought to capture. In addition to remaining in the same lane, drivers also stated that, as they enter a work zone with a temporary concrete barrier to the left, they drive more slowly and cautiously. This change also reinforces the hypothesis that motorists alter their driving behavior in work zones compared to when they are traveling in normal freeway conditions.
|A test driver pilots the instrumented vehicle through the living laboratory.|
Calibrating the car-following algorithms currently used in microsimulation tools with the results from this study will enable modelers to see the effects of behavioral changes caused by work zone activity with greater accuracy. More accurate planning could help transportation agencies better mitigate disruptions on freeways due to work zones, which ultimately will help reduce congestion and delays during peak- and nonpeak-hour traffic.
Up next, the researchers plan to investigate the data and look at the relative velocity and distance (or gap) that each driver maintained on the freeway before, during, and after passing through the work zone. Using statistical software, the researchers will look for trends that will help them better understand how behavior changes due to work zones. They also will test the development of threshold values for a psycho-physical, car-following model using a microsimulation model. Results of this research could be published by the end of 2013.
As Director Joe Peters of the FHWA Office of Operations Research and Development observes, “In a world of increasing value placed upon performance-based management, empirical data on exactly how drivers behave when traversing a work zone are essential for equipping management and operations decisionmakers with the tools they need for deploying the best work zone solutions.”
Taylor W.P. Lochrane is a research civil engineer in FHWA’s Office of Operations Research and Development and is a member of the Transportation Operations Concepts and Analysis team. He holds a B.S. and an M.S. in civil engineering from the University of Central Florida, where he is currently a civil engineering Ph.D. candidate.
Haitham Al-Deek is a professor in the Department of Civil, Environmental, and Construction Engineering at the University of Central Florida. He is the director of the university’s Transportation Systems Institute. With more than 26 years of experience, he is nationally recognized in transportation engineering, planning, and operations fields. He holds a Ph.D. and an M.S. in civil engineering from the University of California, Berkeley.
Jawad Paracha is a transportation specialist with FHWA’s Office of Transportation Operations and leads analysis efforts in the areas of work zone management and traffic incident management. He has a B.S. in civil engineering and an M.S. in transportation engineering and planning from the University of Maryland.
Tracy Scriba is a program manager with the work zone team in FHWA’s Office of Transportation Operations. She leads FHWA’s efforts to implement the Work Zone Safety and Mobility Rule and is responsible for FHWA program areas related to work zone data, performance measures, and best practices. She holds a B.S. degree in systems engineering from the University of Virginia.
For more information, contact Taylor Lochrane at 202–493–3293 or email@example.com.