Congestion Management in the GTHA: Balancing the Inverted Pendulum (Chapter 1)(pages 3-41, Baher Abdulhai, Residential and Civil Construction Alliance of Ontario, Apr. 2013)
Also discussed here: Multi-Agent Reinforcement Learning Integrated Network Of Adaptive Traffic Signal Controllers (Marlin-ATSC)(44 page pdf, Samah El-Tantawy, presentation of PhD dissertation, University of Toronto, 2011)
And here: Making Traffic Smarter - An intelligent transportation system could reduce vehicle emissions in Toronto by as much as 30 per cent(John Lorinc, UofT Magazine, Spring 2013)
Today we review a report on the use of smart traffic lights to reduce waiting times at intersections by about 1/3 which in turn equates to significant reductions in greenhouse gases and toxic pollutants in urban areas. The systems use real-time learning to adjust and optimize signaling.
“Intelligent Transportation Systems (ITS) help increase the effective capacity of infrastructure, manage demand, and maximize efficiency. … consists of a three-pronged approach:
- capacity expansion where warranted,… We can never build enough lanes to completely eliminate congestion at all times and everywhere. Many people see building more roads as an unsustainable alternative.
- demand management to rationalize use.. A little more demand may cause the freeway to slip into a stop-and-go pattern and lose some 25% of its capacity. Preventing this situation by imposing a dynamically varying fee means gaining back 25% capacity at the time we are desperate for it.
- intelligent systems to dynamically enhance efficiency of the existing system, before building more or imposing harsh restrictions on users…A computer system that monitors a freeway and dynamically sets varying congestion pricing fee in real time, such that demand never exceeds capacity, will accommodate 25% more demand.”
“Developed at the University of Toronto, MARLIN is a state-of-the-art traffic signal control system. It is AI-based control software that enables traffic lights to self-learn and self-collaborate with neighbouring traffic lights to cut down motorists’ delay, fuel consumption, and the negative environmental effects of congestion.”
“traffic data collected at 59 downtown locations in 2009, the deployment of a U of T-designed “intelligent transportation system” (ITS) could reduce wait times at intersections by 40 to 70 per cent. As a result, vehicle emissions would drop by as much as 30 per cent.”
“The training (simulation) environment shows us with good precision which intersections benefit the most and how much benefit to expect. This process helps prioritize investment and pick the best candidate intersections (or groups of intersections) to start with.”