Measuring travel-time reliability to reduce traffic delays
Many people deal with traffic congestion on a daily basis, usually during their morning and evening commutes, when it is often predictable and can be planned for. However, unexpected crashes, road construction, or weather conditions can worsen traffic and leave drivers frustrated and late. If this unreliability could be efficiently measured, however, traffic officials might be able to predict or even prevent congestion before it occurs.
Measuring the variability in travel times for any given route—called “travel-time reliability”—has been emerging as a major method by which researchers and traffic officials can quantify the effectiveness of a transportation network.
“Travel-time reliability is another way of looking at congestion and at strategies for making it more tolerable,” says Brian Kary, director of the Minnesota Department of Transportation (MnDOT) Regional Transportation Management Center.
Usually, calculating travel-time reliability involves manually gathering data on weather, traffic, crashes, work zones, and special events and calculating how much a driver’s commute time might fluctuate on a given day in given conditions. However, researchers in the Department of Civil Engineering at the University of Minnesota Duluth have figured out a way to streamline this process.
In a project funded by MnDOT, a research team led by Professor Eil Kwon designed a travel-time reliability measurement system (TTRMS) that automatically calculates travel-time reliability. The system gathers data by automatically accessing MnDOT’s traffic data archive and incident database as well as the National Oceanic and Atmospheric Administration’s weather database. Data for work zones, such as lane-closure periods and locations, can also be inputted.
“It used to take hours, even days, to process travel-time reliability data,” Kary says. “The TTRMS processes it in minutes.”
The system can automatically generate results in both table and graphical formats, thus saving traffic engineers significant time and effort. The TTRMS also includes map-based interfaces, which provide users with flexibility in defining corridors, specifying operating conditions, and selecting types of measures depending on the application.
To put the new system to the test, the research team used the TTRMS to evaluate traffic strategies deployed during the February 2018 Super Bowl in Minneapolis. During the week of the game, MnDOT and the Minnesota Department of Public Safety put extra effort into traffic management to ease tourist-induced congestion. Using the TTRMS, the research team confirmed that this extra effort proved extremely effective; travel-time reliability was actually higher than it had been two weeks before the game despite the increase in tourist traffic.
With insights like this, organizations such as MnDOT can better plan for traffic congestion and make effective changes.
“Since we can’t continually expand the roadways to accommodate traffic,” Kwon says, “the next best method for relieving congestion is to make the traffic system more efficient and reliable.”