TRAFFIC JAMS; PHYSICISTS TURN TALENTS TO SAVING US FROM PERMANENTLY CLOGGED FREEWAYS.

Byline: Lisa Van Proyen Daily News Staff Writer

Scientists long have theorized about the origins of the universe and the species that inhabit it. Now comes the Theory of the Traffic Jam.

Generated by a German physicist and a traffic consultant who have been studying traffic on the mother of all freeways, the Autobahn, they have applied chaos theory and the laws of physics to gridlock.

They came up with the idea that small problems like slow-moving cars coming from a ramp can create monster jams miles away that last hours - even after the problem has been eliminated.

Someday, they say, science could learn so much about traffic jams that thousands of sensors on freeways could feed data into networks of computers that would instantly adjust control devices and avoid that bane of commuting life.

``Even though it's not possible to do this today, it's in our future,'' Boris Kerner, a research physicist at Daimler-Benz in Stuttgart said in a report in The New York Times on Monday.

Kerner and Hubert Rehborn, a traffic consultant from Aachen, theorize in a paper published last week in the journal Physical Review Letters, that road traffic is susceptible to the kind of ``phase transitions'' that water undergoes when it suddenly changes to steam or ice.

``Traffic phase transitions are like cancer,'' Kerner said. ``Once a transition occurs, it may be too late to fix. The answer is timely and accurate forecasting of an approaching phase transition.''

Based on information recorded by roadway sensors, the two found that transitions between a free-flowing highway to jam-packed traffic could occur ``abruptly and spontaneously, sometimes without any change in traffic volume or speed to trigger them,'' the Times reported.

In an effort to computerize traffic information and understand the physics behind traffic dynamics, the U.S. government is doing studies in the Dallas-Fort Worth area and in Portland, Ore.

Officials in Los Angeles say they also are using such high-tech methods to computerize traffic patterns, but they say they have a long way to go.

Physicists at Los Alamos National Laboratory in New Mexico are scheduled to spend $25 million over seven years to look at traffic in the Portland and Dallas-Fort Worth areas.

Eventually, they hope to use the information to create computer simulations that reflect actual traffic patterns and allow officials to find better solutions to problems.

``The beauty of this thing,'' said Richard J. Beckman, a project leader at Los Alamos, ``is that you can try out a policy decision before you actually implement it. For example, you can ask the simulation to show what would happen if you add three new lanes to a freeway.''

In the Portland study, each one of the city's 1.5 million residents will be simulated mathematically in terms of his or her expected activity. That will include information about travel between various points, the likelihood of travel by car, bus, train or bicycle, and other related activity.

``Then you put the whole population together in a microsimulation, run it for 24 hours and see what happens,'' Beckman said.

Los Angeles transportation officials argue that much of the technology needed for computer simulations is already in place or being studied on both the city's surface streets and highways.

``There's nothing new here. There's just a new face they're putting on it,'' said John Fisher, principal transportation engineer for the Los Angeles Department of Transportation. ``It basically (would summarize) what we know now and what we're doing now.''

Fisher cited the Smart Corridor Project as an example.

Stretching about 15 miles, from the East Los Angeles interchange to Centinela Avenue, the Santa Monica Freeway and surface streets are wired with automated detectors in the roadway and changeable message signs to monitor traffic, Fisher said.

More than half of the city's 4,000 traffic signals also are currently on an automated traffic surveillance and control system that measures traffic volume and adjusts the length of the green light by way of computer, said Thomas Conner, general manager of the Los Angeles Department of Transportation.

Conner also anticipates that within a year, high-tech software will be used to anticipate by way of computer what the traffic demands are, rather than reacting to traffic jams.

Currently, the city uses simulated computer models to anticipate traffic patterns, but inputting data is time consuming, traffic officials said.

``There's a tremendous amount of data to put in. You just can't afford to invest that kind of time in this kind of model,'' Conner said.

The city is also about ready to experiment with sensors on buses that give information to a traffic signal to indicate that a bus is approaching and to extend the green light - all in an effort to promote mass transit.

And then there's the National Automated Highway System in the works that features cars that basically drive themselves, with cruise controls that adjust a vehicle's pace and position to more efficiently stack vehicles closer together and reduce collisions.

Though Conner looks forward to technological approaches to battle-clogged streets, he has some concerns about the scientific approach.

``When you're dealing with human behavior, sometimes it's difficult using mathematical models to predict how humans are going to behave,'' Conner said.

CAPTION(S):

Photo

Photo: (color) no caption (road traffic)

Terri Thuente/ Daily News