Toronto: A Model for AV Policy Making

By | August 20, 2018

Benjamin Gillies, Master in City Planning (2018), MIT, 07/28/2018

When Uber first came to Toronto, officials found themselves scrambling to deal with this innovative but disruptive new service. It took them a few years, but after extensive consultation with stakeholders, the City of Toronto would eventually update its municipal code with some of the strongest TNC regulations in North America.

Proactive Planning for New Opportunities

The government’s Transportation Services division learned from this experience: Today, it is working proactively to prepare for the next big transportation disruption, the arrival of automated vehicles. With the Government of Ontario having passed legislation allowing self-driving vehicles on the province’s roads, and Uber already announcing AV fleet testing in Toronto, local policymakers are keen to ensure they get out in front of the impact these vehicles will have.

It was in this context that I spent nine months working with the head of Toronto’s AV working group as part of my MIT Master in City Planning thesis research, looking at how the City of Toronto can effectively regulate automated cars in a TNC fleet. My work examined this topic from a number of different angles, but one of the most exciting areas for exploration was the potential to tactfully employ the unique characteristics of self-driving cars to help address some of the existing shortcomings within the transportation system. Specifically, for the first time ever, it is possible for officials to know in advance what choices a driver will make, so they can contemplate measures that ensure the vehicle’s system acts a specific way, every time, when on the road.

AVs Respecting Local Roads

As an example, we considered was the opportunity to use AV programming to reduce traffic on quiet residential streets. The City of Toronto has designated all its roads as either ‘local,’ ‘connector,’ ‘minor arterial,’ ‘major arterial,’ or ‘expressway.’ City policy stresses that ‚local’ streets are meant for slower speeds and used primarily by local residents, but there is anecdotal evidence TNCs — employing programs such as Google Maps and Wayze — are already taking routes through less-congested local neighborhoods. This increases car traffic on local roads, running counter to the vision of low-traffic neighborhood streets.

When the engineer programs the manner in which an automated car chooses the route it will take, government regulation can prohibit AV-TNCs from selecting local roads unless absolutely necessary, stating that cars must travel on arterials or expressways for as long as possible before approach their final destination. The goal would be to ensure residents enjoy the benefits of living on a local street without having to deal with commuter traffic.

Admittedly, this is a bold measure, and some might question whether it would be easier to employ a road-pricing scheme to similarly deter people from taking certain streets. Yet, if a community lies between, say, the downtown and a wealthy suburb, commuters coming from the suburb might be happy to pay a higher price to take local roads. Is it a resident’s right to be able to enjoy a relative peace on their local road, regardless of whether there are others who would be willing to pay a high price to travel down their street? The current road designation scheme in place in Toronto suggests that it is. As such, the stronger prohibition might ultimately be more appropriate than a market approach, and for the first time ever, automated vehicles give city officials new powers to ensure vehicles respect the road hierarchy.

Of course, this policy is clearly very politically charged, as citizens would likely scramble to have the City designate their street ‚local’ if the government passed such a measure, and not all roads can be labeled as such. Were the government to enact this policy, they would need to refine both the regulation and the road hierarchy plan more broadly. The point of the thesis work was not to say with certainty that this regulation ought to exist, however, but simply to help the AV working group begin thinking about what makes automated vehicles truly distinct from human-driven automobiles, and what new policy opportunities these characteristics unleash.

New Opportunities for All

Similarly, officials elsewhere can look beyond the parameters of traditional travel characteristics, asking how they too can harness the new potential of automation to shift commuter choicesto better reflect local policy. While this would need to be done very carefully and tactfully, such advanced planning could ultimately promote greater equity, sustainability, and vibrancy within their local community.


Toronto: A Model for AV Policy Making was originally published in Harvard Kennedy School Autonomous Vehicles Policy Initiative on Medium, where people are continuing the conversation by highlighting and responding to this story.

Read the original article

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.