Ending traffic jams
Traffic congestion is a daily hassle for many, and authorities are always on the lookout for improved traffic management techniques. But a new partnership between mathematicians Associate Professor Jan De Gier from the University of Melbourne and Monash University’s Dr Tim Garoni with VicRoads may help to decongest our roads.
Dr De Gier from the Department of Mathematics and Statistics at the University of Melbourne explains that urban traffic congestion is a major social, economic and environmental problem.
“Traffic congestion increases travel times, fuel consumption and air pollution and Melbourne’s current urban road network is nearing capacity, particularly during morning and evening peak hours,” Dr De Gier says.
“On urban road networks, traffic signal systems used at major intersections are crucial to determining overall efficiency.”
Dr De Gier explains that most of the current traffic light systems are designed to co-ordinate traffic networks according to a pre-determined schedule, and although this system has worked well over the years, there is a lot of potential for improvement.
“Improvements to the system could include making the signals more adaptive to the actual state of traffic. For example, more detectors in the roads could provide feedback information on traffic levels, and green time wastage (where traffic lights are green but no cars can go thorugh because there is traffic backed up on the other side) could be avoided in directions where there is large traffic build-up,” he says.
But to know what would and wouldn’t work requires lots of trial and error. So the team has designed new traffic models that simulate real life conditions.
Dr Joyce (Lele) Zhang, from the ARC Centre of Excellence for Mathematics and Statistics of Complex Systems at the University of Melbourne has been working with Dr De Gier and Dr Garoni on these models.
“With the help of mathematical modelling and simulations, we are able to investigate things that cannot practically be studied in real-world experiments,” Dr Zhang says.
“Our model allows us to quickly and easily test different settings and analyse their impact on traffic conditions. We can also simulate things like accidents and assess how traffic responds to such scenarios. This allows us to develop a thorough understanding of the traffic network and to identify factors responsible for traffic jams.
“The more knowledge we have about traffic, the better solutions we can find to solve traffic problems.”
The team’s model is novel because it is designed on a mesoscopic level, where hundreds of intersections can be analysed in detail at once.
“Previous models have attempted to analyse traffic on either a small or very broad scale. Small-scale models have good detail but they are restricted to use over specific areas, and broader-scale models lack the detail required to give an accurate assessment,” Dr Zhang says.
“Our model is specially designed to incorporate detail, without sacrificing scale – it is a big step forward in traffic modelling. We can now test strategies quickly, and if they look promising, undertake highly detailed analysis before field testing.”
The team recently received a grant from the Australian Research Council recognising the significance of the research, which will allow them to expand the project in the hope of decreasing travel times, creating safer traffic conditions, and even less pollution.