It has become a cliché to say that we are at the dawn of a technological revolution bringing forward a whole new mobility paradigm. This revolution is happening both on the individual and vehicular levels. As individuals, it is changing our expectations of how mobility should serve us, while on the vehicular level, now that autonomous mobility modes could soon be an everyday reality, it offers the potential to reduce the stress currently associated with driving.

One can imagine the elimination of the stresses related to being stuck behind the wheel in a traffic jam or, since the car can park itself, the stress of finding a place to park. For bus transit, the potential elimination of costs related to drivers could free up money to increase the number of vehicles serving a particular route. What is not to like about these scenarios?

Over the past decade, the primary focus of the research has been on the technology involved to bring about autonomous driving. More recently, there has been research into the impacts of new business models such as ride-hailing or TNCs (technology networked companies) on traffic congestion and transit.

However, during this time there has been very little research into the potential impacts of this evolving mobility scenario upon our physical infrastructure and the public realm. Whereas some are presenting positive scenarios regarding the impacts of autonomous vehicles, others are warning of potential negative outcomes. Either way, it is safe to say that we are at an inflection point, which is likely to be as profound as the arrival of the private automobile a little over 100 years ago.


Proactive vs. reactive

Our current reactive model to the emerging mobility revolution could lead to an inequitable, hands-free version of today's urban transportation framework, whereas a proactive model could realize an equitable and spatially efficient model that the evolving technologies have the potential to release. However, a proactive model requires vision, with goals and objectives being set out by both policy makers and all public realm stakeholders. This model will require making some hard decisions.

On the individual level, our expectations of mobility are changing. With smart phone in hand, we open an app and expect that mobility, in the form of an Uber or Lyft ride, comes to us. We know when our ride will arrive (and can check where the vehicle is as it drives to pick us up), how long the journey will take, and what it will cost. We know the make and color of the car and even the name of the driver. Cashless payment (made after you step out of the vehicle) makes the interaction seem less transactional and more like being picked-up and dropped off by a friend. This is not an accident, but part of the underlying culture of Lyft and its predecessor, Zimride. This model is profoundly re-shaping people’s attitudes to and expectations of mobility. Whereas traditionally the user was expected to accommodate transit’s inflexibility, schedules, and fare structures, today we are expecting mobility to work on our terms.


Quality transit rising

The good news is that the zero-sum impacts that this suggests are not quite turning out as expected. Yes, poor quality transit is being impacted by TNCs, but the ridership numbers for quality transit are rising. And when one looks at the ridership numbers from the GoMonrovia transit program (a partnership between the City of Monrovia, CA, and Lyft), one sees that even within the car-dominant culture of Southern California there can be a positive symbiotic relationship between high-capacity quality transit (LA Metro Gold Line) and the new mobility ecology (Lyft).

On the vehicular level, autonomy offers a hands-free version of what we have today. No stress behind the wheel in traffic, no circling the block looking for a parking space. But is that the best that this technology has to offer? Simply offering a more pleasant driving commute and easier parking would appear to be setting the sights very low and demanding a low return on this massive investment. How can autonomous vehicle technologies achieve higher societal and financial returns on their investment? By combining two of their strengths and leveraging one of their weaknesses (i.e., cost).

Taking on the weakness first, the likelihood of the technology cost-burden per vehicle ever getting low enough to meet acceptable showroom floor retail prices is highly unlikely, so amortization on a cost per mile basis needs to be part of the solution. This can be achieved through fleet ownership by OEMs, with consumer access on a subscription basis. The beta testing of this ownership model can already be seen in examples such as the Volkswagen Group’s Audi on-demand “Own the experience, not the car” campaign.

Further, with a fleet-ownership/subscription-access model, the current 4% utilization rate of private autos in the U.S. could potentially be inverted to 96%, making the utilization and deployment of high-capital-cost AVs more closely resemble that of the airline industry. In addition to amortizing these higher capital costs over more miles, subscription access would reduce the overall fleet size required to serve a given population as a vehicle is no longer serving one owner but multiple users instead. Estimates by the Boston Consulting Group and MIT have cited potential fleet size reductions in the range of 60% to 78% for such a shared autonomous vehicle environment.

When we combine the potential for significant fleet-size reductions noted above with the inherent spatial control of an autonomous vehicle, through reduced lane-widths and closer headway between platooned vehicles, one can accommodate more vehicle operations per square foot of pavement. This can lead to two outcomes; either increase the vehicle throughput capacity of a given right-of-way, or keep the throughput at its present level and assign the efficiency benefits to active transportation or other non-vehicular uses instead.


The case for multiple modalities

In separate studies carried out by Perkins+Will, with Arup and Nelson\Nygaard, it has been demonstrated that in an AV scenario the overall vehicular right-of-way can be reduced by around 40% while maintaining the same vehicular throughput. However, traditional vehicle throughput metrics have a bias toward the single occupant vehicle in that they do not differentiate between a solo driver and a 60-passenger bus. When street efficiency metrics are revised to reflect throughput measured by the number of people per hour, there is a radical change to the street design and its efficiency.

In a Lyft-commissioned study in 2016 by Perkins+Will and Nelson\Nygaard of a typical Los Angeles boulevard (Wilshire Boulevard in the Westwood district was chosen to be representative), the current street throughput of people per hour was 29,600, relying predominantly on SOVs (single occupancy vehicles) and limited transit. However, with predominantly shared AVs/TNCs and high-frequency/short-headway autonomous bus rapid transit (BRT) modalities, and incorporating enhanced dedicated active transportation lanes, the throughput of people per hour increased by 260% to 77,000.

Earlier it was noted that the current reactive stance to the development of these exciting new technologies will likely lead to benefits confined largely to the occupants of privately owned or leased AVs. This is the low return on social and financial investment scenario; essentially a hands-free version of the status quo.

However, the studies noted above, using an ownership model that more realistically addresses the realities of the costs involved in deploying these technologies, illustrate a win-win situation. They illustrate both an enhanced public realm as well as a more efficient pavement that does not prioritize a single modality, but instead recognizes all public realm stakeholders within the right-of-way.

This outcome will require leadership and thoughtful planning involving input from all who use the public realm and not relinquishing control to one particular group of users. This approach looks for a higher social and financial return on investment and reflects the transformative potential of these technologies not just for a minority, but for everyone who uses the public realm.

To understand how achievable this could be, even in southern California, the numbers coming out of the GoMonrovia program indicate that, under circumstances tailored to meet their needs, people may be more willing to forgo using their own car than we think. For instance, ridership has gone from 4921 in March 2018 to 75,808 in January 2019, and it currently exceeds 77,000 riders per month, of which 90% were shared rides.

We are at an inflection point; we can do nothing and remain in our current reactive mode, or we can start the discussion now to set out the vision for the future that we all want. Look at Wilshire Boulevard today or look at the image above; which would you choose? This is more achievable than we realize, but first we need to be at the table and participate in the process.