On the Road
Addressing Traffic Infrastructure Demands
In all of history, no society has witnessed urban growth and development at such a rate. While occupying only three percent of earth’s land mass, cities are home to over fifty percent of the planet’s population. By 2050, that percentage is anticipated to reach almost seventy percent with a burgeoning population of ten billion according to the United Nations.
The new reality is that the future is one of urban living, and Canadian cities are not exempt. With a population of over thirty-five million and increasing, this country is the fastest growing in the G7, according to Statistics Canada, with Montreal, Vancouver, and Toronto being the largest metropolises. This is creating a problem for traffic management in these cities. Cities will have to seriously plan to move populations in an efficient, practical, and safe manner, with a focus on avoiding traffic congestion – every driver’s nightmare.
“Any large city or a metropolis where demand for transportation is higher than the capacity of the infrastructure, agonizes with congestion,” says Professor Baher Abdulhai, director at the Centre for Automated and Transformative Transportation Systems (CATTS) and Toronto’s ITS (Intelligent Transportation Systems (ITS), Department of Civil Engineering, University of Toronto. “This is universal, all over the world.”
All large cities are affected and, “to a lesser degree, the smaller cities. Small cities have small infrastructure. So there too, there is congestion,” he explains. “When demand for transportation is higher than the size of the infrastructure, then congestion mitigation is relevant.”
The congestion issue, “grows gradually,” says Professor Abdulhai, “but it gets to the point when it becomes unbearable.” He notes that everything has to be measured, “relative to free-flowing systems.”
He uses, as an example, a driver travelling from point A to point B at midnight, who will not generally anticipate any traffic problems. This driver is commuting at full speed with a free flow of traffic and arrives at the destination with no delays. But if this driver is commuting at peak periods, as many do, it will take the driver three times as long to get to his destination. “This is a serious issue. That makes the stress of traffic relative,” he says.
“It makes sense to measure what is happening in a given city relative to what people wish for with no congestion. Not that free flow is necessarily desirable in terms of efficient infrastructure utilization, but it’s what people imagine and wish for,” he explains.
“When demand over supply (number of roads, size of public infrastructure, for example), approaches or exceeds one, then we have a problem … not everyone can go at the same time without significant congestion,” Professor Abdulhai explains. And since most individuals have to travel at the same time to get to places of employment, the solution would be, “either reduce the demand or increase the supply capacity.”
The question then becomes – how do we do that? He suggests three categories of solutions, one of which is to build more roads or public transit systems. The debate as to which option is most feasible remaining a pressing issue. “But, collectively, that’s building transportation infrastructure,” he says. “That’s a solution, and it’s inevitable.”
He notes some drawbacks to this solution as it is very capital-intensive and infrastructure is very slow to construct, even with available capital. Also, land availability may be an issue for some cities, so there are constraints regarding budget, timeframe, and space. And certainly, environmental impacts have to be taken into consideration.
Another solution may be in managing demand. He questions the sustainability of most vehicles having a one-person occupancy. Carpooling may be an option to, “consolidate demand and put as many people as possible in one vehicle.” Additionally, social taxis like UberPool may get “as many people as possible into a car.” And certainly considering and encouraging public transit is viable since it represents, “the biggest form of carpooling.” Employers can also encourage the practice.
“All of these tactics or methods are focused on reducing demand … If every two drivers of two cars carpool, then we can cut down demand by half. That can go a long way. If we focus on the simple equation which starts with demand over supply capacity exceeding one, [we] can significantly reduce demand while attempting to build or supply infrastructure, whether it’s transit or roads.”
An issue with these forms of management is that “It’s usually inconvenient,” explains Professor Abdulhai. “Many people don’t like that and resist it. One method that is picking up momentum, worldwide, is congestion pricing, which is controversial.”
Congestion pricing is a toll that applies when driving into areas that are heavily overcrowded. This may not prevent everyone from driving into congested areas during certain times of the day, but that is not the goal. “The goal is to reduce demand a little bit so that it would fit in the existing infrastructure capacity,” he says. “It’s a disincentive to rationalize demand.”
He notes that other cities across North America control access to freeways with ramp metering, which is a form of controlled access using a traffic light. “It’s used all over the world. It’s used in Australia, the United States, and in Europe. In Canada, we’re a bit behind.” These solutions need to be implemented concurrently to be effective.
The third solution bundle comes in the form of using efficient advanced technologies. Professor Abdulhai believes that our current traffic light system functions in its simplest form, but there are ways to make lights smarter and actively engaged in traffic conditions. His most recent work at the University of Toronto is involved in deep learning systems for traffic lights. “A traffic light can have eyes like cameras or radar technology and a brain to think and act upon what it sees – to adapt to traffic conditions – second by second.”
There are also vehicles that can communicate with traffic signals via vehicle-to-infrastructure (V2I) technology in which a traffic light foresees approaching traffic. So, in addition to possessing ‘eyes,’ it also listens to what is coming. It can be conscious of traffic flow approaching and can use intelligence, “to decide when to switch and for how long,” he says. “Continuously, the traffic light is on its tippy toes in order to be as efficient as possible.”
Such technology has been evolving for more than twenty years. Toronto had an earlier system called SCOOT® at some intersections for decades. “However, like smartphones, technology advances very rapidly, and twenty-year-old technology is ancient,” he adds.
“Compared to a basic traffic light that we all know, it can have up to fifty percent less delay,” he says of such traffic light technology. If, for example, a commute trip is thirty minutes and one passes through twenty traffic lights, a smart system can decrease the delay by ten minutes. “This is substantial.”
With technology such as intelligent transportation systems, “The infrastructure can take more demand without building more infrastructure.” Such solutions enable us to, “be as intelligent and efficient as we can and manage both demand and the operation of the infrastructure concurrently.”
The introduction of the first autonomous vehicles was in the 1980s, and the technologies used in these vehicles has been steadily evolving, but not without some controversy. These vehicles have been, “touted as a solution, and they can be,” says Professor Abdulhai. This is both convenient and results in fewer accidents. “Most accidents happen because of human error. This automated chauffeur can exacerbate our dependence on the car, induce more demand and create more congestion.”
On the other hand, since these vehicles are driven by computers, computers can also solve that issue. “More of these vehicles can fit in the road and increase the capacity of the road.” He explains that if a given freeway can handle two thousand drivers per hour, a properly automated vehicle could double the capacity without the need for new infrastructure.
However, autonomous vehicles, at this time, leave sufficient room from the car in front so that the computer can understand and react to traffic flow, “which effectively reduces the road capacity; the opposite to what we hoped for.”
There are also risks and unintended consequences of convenient autonomous vehicles. “Obviously, cybersecurity and the potential for hacking is one of them,” continues Professor Abdulhai. “More individuals will use cars, which is the opposite of what we’re trying to do, which could potentially increase demand.” Also, since the daily chore of driving is gone, some people may choose to live further from cities, “which could increase urban sprawl.”
So while autonomous vehicle technology may bring hope, “There’s also equal amount of risk and unintended effect. We’re trying to better understand the true impact of technology like this and advance traffic management to exploit the pros of technology while avoiding the cons,” he says.
“We have to regulate and manage the system in order to take it in the right direction,” he says. If demand spirals out of control because of this technology, “then we need to use the technology itself to manage demand.”