DETROIT, Mich. -- WSU researchers are working with the Michigan Department of Transportation to make your driving experience smoother and safer.

The team of students continues to analyze a merge system new to Michigan using traffic sensors and flashing yellow lights, successfully tested for the first time in metro Detroit last summer at a construction site alongM-53 in Macomb County.

The pilot study began two summers ago at construction zone sites along the interstate highways in Bay, Eaton, Branch and Muskegan Counties. After refining the system, the team returned last summer to conduct tests along M-53 in Shelby Township and I-96 in Grand Rapids.

Roughly half the 7,000 traffic accidents in construction zones in Michigan in 1999 occurred on roadway and lane closure areas, says Tappan Data, professor of civil engineering and director of the College's transportation study group. With the new merge system, drivers start merging well ahead of time, creating a more ordered and less congested flow of traffic. "We see a significant decrease in the number of drivers who insist on passing others up until the very last instant before jumping in," he says.

Last July, a Detroit News reporter visited the M-53 test site in Shelby Township where Michigan State troopers issued traffic fines for drivers who ignored the flashing 'no passing' signs or speeded through the zone. The troopers have positioned themselves at times at all the sites to help drivers obey the new system.

The merge system, named Dynamic Lane Merge Control, is getting wide publicity, which is good, according to Colleen Hill, and co-leader of the project. The more drivers familiar with it, the better it works, she says. "There's a learning curve involved. Once they've gone through it, they understand better what they're supposed to do."

The merge system uses a series of left (or right) lane 'Do Not Pass When Flashing' signs mounted on trailers and placed 1,500 feet apart, explains Hill, a graduate research assistant. The signs are linked to electronic sensors and microprocessors that detect the volume of vehicles moving through. When the volume rises to a pre-determined threshold toward the head of the merge zone, a signal is sent to the next sign down the line to start flashing. When further congestion is detected, additional signs downstream activate to stretch the line of cars even more.

Kerrie Schattler, the other project co-leader, says she expects the new system to be used at more two-lane construction sites in the state. "The system as a whole is going over very well. I believe MDOT intends to use it in the future. It's very effective." The system is most practical along two-lane highways with moderate traffic at peak hours, she adds.

The WSU study is being conducted under a $101,000 MDOT grant. The initial design of the system was based on the Indiana Department of Transportation merge system. After MDOT selected the sites and implemented the initial system designs during the summer of 2000, it asked Data to analyze the system's effectiveness and recommend improvements.

The first summer, Hill and Schattler traveled extensively to the four far-flung counties, setting up at each site to collect data on vehicle volume and driver behavior.

Last summer, the team collected data at the M-53 construction zone site in Shelby Township before the merge system was put in place there, as well as along I-96 near Grand Rapids. This allowed for a baseline comparison to test the effects of a merge system the team fine-tuned in the lab. Besides refining the sensor relay timing, they made additional adjustments in the distance between the signs and added 'positive guidance' signs to help drivers obey the system.

To determine the sensor occupancy settings, they relied on a textbook formula based on distance traveled, average speed and flow rate, modifying it to remain accurate to the varying conditions at each test site.

The team improved the average 85th percentile speeds through the M-53 work zone from 56 mph to 67 mph, says Data, presumably due to smoother traffic flow created by the merge system. The average travel time delay per vehicle through the zone decreased from 254 to 164 seconds. The team also recorded a significant reduction in aggressive driving moves, although that may have been due in part to the presence of state troopers.