News

This blog is based on upcoming UKERC research.

Autonomous Vehicles (AVs) may be key to reducing carbon emissions from land travel. Transport emissions are flatlining but not coming down, and AVs are seen as a techno-fix.

Research to date says carbon savings from AVs will mostly come from sharing them. This is because they could ferry people all day instead of sitting on private drives, reducing the number of vehicles required.

So, we were very keen on exploring views on shared AVs; in car clubs, or as taxis or buses. People do not use such shared vehicles much now – so would AVs be any different?

What we investigated

We investigated the public views in three ways:

• A choice experiment, where people were offered direct hypothetical choices between trips on current or autonomous vehicles of different types; and
• Deliberations – focus group-style guided discussions about what an AV future would look like; followed by
• An MCDA (Multi-Criteria Decision Analysis) exercise – a method of getting people to assess different options. It involves giving the options scores using criteria of different importance (or weight).

By using these three methods, we hoped to avoid a common issue in research: thinking that one method or theory can reveal everything. The choice experiments used a representative sample of the public, the other two methods also sought out non-typical views.

The choice experiment offered direct hypothetical choices between transport modes for trips with varied costs, times, and convenience. The choices were converted into a Value of Travel Time Saved (VTTS), by which modes can be compared. It is assumed that the cheapest, most convenient option will be most popular. The experiment also measures the importance of factors that it didn’t present to people on their choices. These ‘imponderable’ factors might include:

• The technology itself (e.g., embedded carbon, energy efficiency);
• The services it provides (e.g., travel and wait times);
• Psychological factors (e.g., trust, anxiety, safety, etc.), familiarity, or the rejection of new technologies, etc.

The deliberations asked people about 100% AV futures:

• How would they differ from today?
• Were they feasible or likely, and if not, why not?
• What might slow or speed up transition to such a future?
• Are shared or private AVs more appealing, and why?

Open-ended questions like these revealed detail of the ‘imponderables’. These include social, cultural and political issues, and also issues like trust, concerns, or subjective views based on age or gender.

The MCDA exercise then led people through assessing different options. Although we allowed people to assess different vehicle fleets, almost all compared transport modes. These were conventional and autonomous, private or shared, as the public chose. They also selected their criteria, how important they were, and the scores. This produced a ‘rational’ assessment that often differed from what they themselves expected; car enthusiasts often scored buses highest!

We found that our three methods – perhaps predictably – came up with different insights. These highlight different sides of the issue, just as asking different questions gets different replies.

The choice experiments found that beyond service attributes (i.e. convenience), they were preferred less than conventional cars. The deliberations raised 50% more concerns than benefits of AVs, and yet the MCDA assessments mostly scored AV options highly. Shared AVs came out generally as most preferable, although there were fears about unmanned shared taxis and buses.

Another key finding from all three activities was that views of shared AVs seem to be based largely on how buses, taxis, and EVs are currently experienced. If these modes of transport currently fail people, perhaps they need to be sorted out for AVs to have a future?