Adelaide study suggests driverless cars could lead to more traffic congestion partly because of drivers’ attitudes
24 October 2019
New research from the University of Adelaide suggests that driverless cars could worsen traffic congestion in the coming decades, partly because of drivers’ attitudes to the emerging technology and a lack of willingness to share their rides.
Using the City of Adelaide as a test model, researchers surveyed more than 500 commuters, including a mix of those who travel to work by car and public transport, and modeled the potential impacts. The results are now published in the journal Urban Policy and Research.
Autonomous or driverless vehicles are likely to have profound effects on cities. Being able to understand their impact will help to shape how our communities respond to the challenges and opportunities ahead.
—co-author Dr Raul Barreto, from the University of Adelaide’s School of Economics
The multidisciplinary research—conducted by the University’s School of Architecture and Built Environment, School of Economics, and the Australian Institute for Machine Learning, in collaboration with researchers from the City of Adelaide—investigated commuters’ views on autonomous vehicle ownership and use, vehicle sharing, and their attachment to conventional vehicles.
The research team then explored potential vehicle flow, with a mix of autonomous and conventional vehicles, and land use change in the Adelaide CBD under different scenarios.
Our findings show that Adelaide has the potential to significantly reduce the number of vehicles on the roads and improve traffic flows, however these benefits may not be achieved in the near to medium term for many reasons. The key factors affecting the transition to autonomous vehicles are commuter attitudes to car ownership and wanting to drive themselves, rather than have technology do it for them, as well as the price of new technology, and consumer attitudes to car sharing.
Our evidence suggests that as riders switch to autonomous vehicles, there will be an adverse impact on public transport. With most commuters not interested in ride sharing, this could increase peak period vehicle flows, which is likely to increase traffic congestion over the next 30 years or so.
Under both scenarios we tested, the number of vehicles overall will eventually drop. However, total vehicle trips may increase, and some of the predicted benefits of autonomous vehicles may not eventuate until a lengthy transition period is complete.
—Dr Barreto
Resources
Jon Kellett, Raul Barreto, Anton Van Den Hengel & Nik Vogiatzis (2019) “How Might Autonomous Vehicles Impact the City? The Case of Commuting to Central Adelaide,” Urban Policy and Research doi: 10.1080/08111146.2019.1674646
Shared e-mini buses may be a way to restrict traffic increases?
Posted by: HarveyD | 24 October 2019 at 06:52 PM
Of course they will create congestion, If people can do whatever they like in their cars, instead of driving, they won't mind spending longer in the car.
They could live further away then they do now and sleep, or work or surf while being driven in for 2-3 hours, no problem.
Also, if the car could drop them at the office and then park elsewhere, what do they care if it clogs up the traffic.
+ the notion of sequential car sharing is BS, you might get 2, maybe 3 goes each morning, but that is it, everyone wants to go to work at more or less the same time.
As Harvey says, you have to get more than one person into each car. Ideally 4-6 in a minivan.
I heard recently that people prefer to car share with 2 or more extra people, rather than just one. One is a bit intense.
Google should organise car sharing, or at least facilitate it.
They know your every move, where you live and where you work, and what time you go there. They could easily pair you up with others with similar or overlapping routes and build up little "daisy chains" of commuters.
You'd have to solve the privacy problem.
IMO, opting in is not enough, because most people wouldn't. You would have to tell them that there are good options for them to car share (if they are in a daisy chain).
Once you get them to opt in, they can input their preferences, (say talk/no talk, classical / pop music, democrat / republican, man/woman).
If you think of it, there is a huge gap between single car occupancy and public transport, where you could share with 100 people.
If you can find a socially acceptable way of bridging that gap, you have solved the traffic problem.
Posted by: mahonj | 25 October 2019 at 12:24 AM
I fully agree that car/mini-bus sharing, when well organized, has a great future, specially with driverless e-vehicles. It could greatly reduce city cores trafic jams, pollution and GHGs.
Posted by: HarveyD | 25 October 2019 at 06:50 AM
lack of willingness to share their rides....
That is the key point.
We could have ride shared since the oil embargoes of the 1970s but NO.
Posted by: SJC | 26 October 2019 at 02:55 AM
It may be a matter of better management and automated applications on cell phones. If Amazon can manage a ONE DAY delivery for millions of articles, UBER (and similar organisations) could certainly manage shared rides with minimum delays and more so with automated driverless vehicles.
Posted by: HarveyD | 26 October 2019 at 08:59 AM