Science and Technology

Smart shared mobility and potential implications for levels of congestion

mobility through the looking glass
Photo by Bekir Dönmez on Unsplash

In the second of a series of six articles, Professor David Hensher of the University of Sydney discusses whether shared transport options will indeed reduce congestion, and the roles of autonomous vehicles and public transport in the new mobility ecosystem.

The transition to smarter mobility that is taking place, referred to as Smart Transition, typically involves greater car sharing (facilitated by apps) and less owning of cars by private individuals, as well as the future role of (electric) autonomous vehicles. It has an underlying mandate to redefine and commit to a Collaborative and Connected Society (CCS) whereby the mode is far less important that the service levels that satisfy customer needs.

While we will always need reflective and effective governance frameworks to ensure deliver of CCS, we have an opportunity to finally break the stranglehold that outmoded mode-specific regulatory models have had on the provision of transport services1. Why should we continue with mode-specific contracts, often associated in the public transport sphere with public monopolies or provided by competition for the market (i.e., competitive tendering – see Hensher 2017), all supported with provider-side subsidies? This includes the limitations imposed on the overspecification of network service levels (and the predominance of timetables for conventional public transport).

The car-based systems associated with taxis are now being broken by the arrival of new service models such as Uber and Lyft, although they are essentially mode-specific (though covering an increasingly expanding mix of intermediate modes, many of which are being defined for the first time). Fundamentally, we increasingly see many variants on the conventional wisdom that are tantamount to delivery models that cannot operate under outdated regulations. Smart Transition is the context in which we have to contemplate that anything goes as long as it has a sensible customer outcome, and one might hope an acceptance by government as the custodian of societal interests through a reformed governance (and funding) model.

Will collaboration reduce congestion?

The current interest is in how this all relates to the future of road congestion reduction, something that is claimed to be a major benefit of an era of intelligent mobility. This appears to be premised on one crucial consideration, the success in moving society to a regime of collaboration and connectivity, initially without autonomous vehicles but subsequently with such vehicles.

Collaboration is often associated with the sharing economy which can take at least two paths – shared and pooled (see Wong, Hensher and Mulley 2017), or without others, for a particular ‘point-to-point’ or ‘point-via-another-point-to-point’ trip. It is far from clear how much of the congestion challenge can be resolved through greater sharing of private cars (no matter whether they are autonomous or not), increasing occupancy, assuming a constant number of person trips. However, sharing of private cars could lead to increased trips overall through a higher number of trips per vehicle, and to greater congestion if the number of trips overall goes up.

A very specific issue being raised within the new reform agenda is what all of this might mean for the number of cars on the road and the amount car usage (vehicle kilometres travelled). The limited evidence on smart transition (predominantly associated with smart apps, opportunities to ride hail and dispose of a car), is simultaneously creating the promise of a system that can reduce demand (congestion), but at the same time fulfilling previously unmet demand and creating new demand. Smart transition moves society to a rentier model where the incentive for the mobility service provider is to generate as much mobility as possible (i.e., trips and kilometres) to maximise returns on capital (Karlsson et al., 2016).

What little evidence there is at present, based on simulated scenarios of futures, is informative, but can it be relied on? Two studies are of particular interest, one from the ITF/OECD, and one from University of California Davis (Clewlow and Mishra 2017). These studies say nothing about the impact of autonomous cars or indeed any renewed future role of public transport (except on-demand buses) – they primarily focus on shared cars with a driver, the latter likely to be the basis of car travel for at least the next 20 years.

The ITF/OECD study modelled the impact of replacing all car and bus trips in a city with mobility provided through fleets of shared vehicles. The study found that if all individually-owned private cars were removed from the city with shared vehicles only, there would be a substantial reduction in the number of vehicles required to service overall mobility demand, and greater equity of service across the city as a whole. However, the findings suggested an increase in vehicle kilometres driven of 6.4 percent per day.

Once the assumption of perfect conditions breaks down, and 50 percent of private cars are assumed to remain, the performance of the system deteriorates further, with up to 90.9% more kilometres being driven per day. This does not sound like a congestion buster? Even more congestion on our roads; although the congestion levels may be more predictable (non-random) with improved reliability, and maybe a lower value of travel time savings and reliability willingness to pay.2

ITF/OECD (2017) also undertook a simulation study, using mobility and network data from Lisbon, Portugal and examined scenarios where shared mobility is delivered by a fleet of six-seat vehicles (shared taxis) that offer on-demand, door-to-door shared rides in conjunction with a fleet of eight-person and 16-person mini-buses (taxi-buses) that serve pop-up stops on demand and provide transfer-free rides. Rail and subway services are assumed to keep operating in the current pattern.

They tested scenarios where car owners could use their car for one, two or three days each working week, which corresponds to having 20 percent, 40 percent, and 60 percent of trips currently made by private car continuing to be made by that same mode3. Allowing for 60 percent of the private cars4 brings virtually no reduction in congestion, thus producing no visible result of improvement, and no support for the political argument in favour of introduction of shared mobility solutions. On the contrary, allowing 40 percent of the private cars (each private car allowed two days per week) reduces vehicle kilometres at the peak hour by 13 percent, which according to the authors ‘essentially makes congestion disappear’.

Findings from seven major US cities

The reduction of parking space needs is also visible (see Rantasila, 2016) and it is suggested that this would allow a benefit for pedestrians (and cyclists) in many parts of the city (assuming this space is not used for other endeavours). Environmental impacts are also likely to positive (Brendel and Mandrella, 2016). These “quick wins” can be essential to gain political support for the change.

Clewlow and Mishra (2017) presents findings from a comprehensive travel and residential survey deployed in seven major U.S. cities (Boston, Chicago, Los Angeles, New York, San Francisco/ Bay Area, Seattle, and Washington, D.C.), undertaken in two phases from 2014 to 2016, with a targeted representative sample of urban and suburban populations. 4,094 completed responses were collected between the two surveys, with 2,217 from respondents residing in dense, urban neighbourhoods and 1,877 from more suburban locations5. They show that directionally, based on mode substitution and ride-hailing frequency of use data, that ride-hailing is currently likely to contribute to growth in vehicle miles travelled (VMT) in the major cities represented in this study6.

Specifically, they find that ride-hailing users, on average, do not possess significantly fewer vehicles than their non-ride hailing counterparts, and have more vehicles than those who only use transit. While some amount of ride-hailing users reduce the miles that they personally drive, and nine percent disposed of a vehicle, these miles return in the form of miles travelled in a ride-hailing vehicle. They found a strong correlation between increasing ride-hailing use and increasing rates of vehicle reduction. That is, the more frequently an adopter uses ride-hailing services (from once a month to daily), the more likely they were to have reduced their household vehicles.

The reduction of vehicle ownership is primarily of value insomuch as it reduces total vehicle miles travelled (VMT), although the reduction in the number of vehicles is the system is encouraging. What is currently unclear is the net vehicle miles travelled (VMT) adjustment due to the introduction of ride-hailing – has it gone up or down? And what are the likely longer-term impacts of these services as ride-hailing companies operate a commercial model with a usage maximisation objective?

The role of autonomous vehicles

There is nothing in this evidence to inform us about the role of autonomous vehicles. Whereas the studies report above are related to driver based cars, a study by MacKenzie et al. (2016) estimates that autonomous cars can cut the cost of travel by as much as 80 percent, which in turn drives up kilometres travelled by 60 percent7. Clearly the price elasticity is at work and really does matter. MacKenzie suggests that “You are talking about a technology that promises to make travel safer, cheaper, and more convenient. And when you do that, you’d better expect people are going to do more of it.”

As interest in shared mobility grows, with evidence thus far mainly focussing on switching between a privately-owned car usage model to a car-based shared vehicle model, a broader question of interest becomes the mix of modes that will (might) be offered through the subscription plans that are starting to enter the market, or being actively considered, by mobility service providers (and how these ought to be allocated including the link with modal efficiency and land use—see Wong, Hensher and Mulley 2017). When autonomous vehicles are added into the mix in future years, and there is greater acceptance of sharing non-owned cars (in contrast to sharing existing public transport)8, there are real prospects of significant changes in the performance of the transport network, especially roads9.

But the question remains – how much of the kilometres travelled on the roads by shared vehicles will be car based and result in significant reductions in conventional public transport use (Hensher 2017)? Will this also make traffic congestion worse, even if more predictable, but in a safer mobility environment, or not? Maybe the advent of the small bus will become nothing more than a higher vehicle occupancy shared car, which maybe a good thing and is almost inevitable that it will become the bus mode under a demand-responsive system10?

The importance of public transport in the new transport ecosystem

MaaS (Australia)11, for example, argues that without public transport at the centre of MaaS complemented with on-demand transport covering the first and last mile, it will not succeed in delivering mobility solutions that are more than simply commercial propositions designed to make money for car-based services, but will deliver on the goals and expectations of government and society more broadly. It is a balancing act – optimising the supply and demand chains of the transport services without compromising the intended goals and objectives. This will require all transport providers and MaaS operators to collaborate transparently (the commercial model yet to be contested / trialled).

This creates a real challenge for a number of the players in this space who have a vehement interest in a car-based solution, and who are reticent about joining forces in a subscription plan with conventional public transport modes, which they believe will be loss making. This raises important questions about how the packages can be designed to allow for internal cross subsidy12 and still deliver an acceptable return on investment, while at the same time not requiring government subsidy, and hence allowing the MaaS model to be delivered under a truly free market setting. MaaS (Australia) envisions that mobility ecosystem benefits will be realised by establishing a trusted partnership between the public and private sector and unlocking the access to data.

The ‘limited’ evidence also suggests that if the cost of mobility drops under a CCS model (after accounting for removing ownership and hence costs, and eventually the widespread presence of autonomous cars), then we should expect more road-based travel (i.e., increased passenger kilometres), but what is unclear is whether the move to sharing will reduce vehicle kilometres – it may be fewer vehicles doing more kilometres since they will be on the road for many more hours (being more temporally efficient) than the privately owned vehicle. If you believe some pundits, this is the model that will evolve as the business model of the growing number of competing ride-sharing and ride-sourcing services, forcing prices down13 and increase the hours that any one car has to be on the road to recover the commercial objective (aligned with arguments given by Hollands 2015). Conventional public transport needs to heed this warning as does government.

Part of the solution in protecting the future of public transport may rest with future government funding models14, especially if the current provider side subsidy regime for public transport is replaced with a user-side controlled scheme15 that gives users control on where subsidy is allocated given what public transport trips they choose. This will also engender a competitive process as mobility providers compete to attract customers, and hence is expected to result in a much more efficient level of subsidy provision and hopefully modal mix of offers.

If it results in reduced demand for public transport services and increased demand for shared car services, government will have to decide how much of conventional public transport should continue to be provided, both in the MaaS package and separate from such packages. It is likely that conventional rail (and bus rapid transit) will remain as the backbone of the transport network (even if scaled back where services are no longer justified), but the rest of the public transport system (essentially bus based and increasingly demand responsive) might be folded into the new mobility model.

In summary, will MaaS schemes be dominated by car-based offers? What does it mean for conventional public transport, and government’s response to a car dominated solution? Does it matter what modes are in the mix? Maybe in the future the autonomous car in a safe platoon will act like public transport and so does it not matter what the mode is? This, however, is likely to be less spatially efficient than purpose-built large vehicles (and higher cost autonomous or not, being unable to reap economies of scale from shared components—engines, etc.) and possibly sabotage many of the intrinsic benefits (e.g., land use) of fixed route mass transit.

The land use implications (in terms of density and where people will live and work) must be given closer scrutiny since the initial transition to sharing under MaaS without autonomous vehicles, and the subsequent roll out with a significant amount of autonomous vehicles, will change the location landscape, with even more diverse origin-destination patterns and longer commutes expected. The one instrument that can protect a preferred land use and activity profile for our cities in particular is pricing and funding reform, which may be the greatest disruptive instrument available to government to achieve its desired objectives. This should be central to the new governance model and an opportunity not to be missed.

Footnotes

1. Interestingly, NSW made a start on this with the 2014 Passenger transport Act, where linking of specific vehicles of modes was removed.
2.The study only forecasts a minor increase in travel times by having distributors and local streets absorb much of this increase—thereby ignoring the road hierarchy and bringing associated externalities.
3. A related issue and cautionary evidence is what happened when Athens allowed vehicles to be driven every other day (odd and even registration plates), albeit for pollution control. The outcome was an increase in the number of vehicles as households moved to buy more vehicles so they had a permitted vehicle for each day or they went for one of the exempted vehicles. It is not easy to stop people once they have a car, so reducing the number of vehicles overall has to be where we start.
4.The study found that if conventional cars were replaced with driverless cars that take either a single passenger at a time or several passengers together, as long as half of travel is still carried out by conventional cars, total vehicle miles travelled will increase from 30 to 90 percent, suggesting that even widespread sharing of driverless cars would mean greater congestion for a long time.
5. The authors correctly express concern about many other studies in the USA that have self-selection bias in that their samples are from locations where car ownership is typically much lower than normal, and the socio economic characteristics and density of the built environment are more supportive of not owning a car and using taxi-like services.
6. A recent US Department of Transport study of autonomous car sharing in Chicago predicted a 70 percent worsening of traffic congestion. 30 percent is the cited figure for San Francisco. Referenced in McCabe (2017) video talk at http://ozebus.com.au/information-for-moving-people/confspeakers
7. To realise this savings, significant investment is required up front with a long term ROI model.
8. This also implies a greater level of personal intimacy–how acceptable this is has yet to be adequately tested.
9. All this depends on the pricing model. One of the serious impediments to sharing is the way in which the private car is a low marginal cost mode, especially in comparison to shared cars, public transport etc. MaaS relies on people being prepared to understand the full costs of car ownership.
10. An important question is to what extent conventional fixed route timetabled bus services will be replaced by demand-responsive bus services (DRTs), and importantly what this will mean for future large bus needs and contracting models. Some pundits have suggested that DRTs can be built into existing contracts (as variations) under tendered or negotiated contracts; however the author argues that DRTs should be part of an economically deregulated market, at least initially, as a way of encouraging competition. If as might be expected in some jurisdictions, the DRT appeal is in replacing conventional (RPT) services with DRTs because of the thinness of the market, then a natural monopoly setting may evolve which suggests that a single operator makes sense. But this does not have to be the incumbent RPT provider, especially where the buses are small and not available under a shared cost model with the incumbent RPT services.
11. I thank Hany Eldaly (Managing Director and Co-founder of MaaS (Australia)) for discussions on this theme and the commitment that MaaS (Australia) has to ensuring public transport is in the mix. See http://maasaustralia.com/
12. This is the basis of a current study by ITLS, led by Corinne Mulley, on community transport in NSW.
13. Including low labour costs for drivers, which is a criticism of many Uber services; and with driverless cars in the future these costs will be even lower (possibly halved), making car based travel very popular at the risk of a future role for public transport.
14. I am indebted to John Stanley for suggestions and discussion on this point.
15. As is potentially going to happen for with the community transport sector in New South Wales.

Author: Professor David Hensher

Professor David Hensher is Founding Director of the Institute of Transport and Logistics Studies at The University of Sydney. Among his many recognitions and achievements, David is a Fellow of the Australian Academy of Social Sciences, and the recipient of the 2009 International Association of Travel Behaviour Research (IATBR) Lifetime Achievement Award in recognition for his long-standing and exceptional contribution to IATBR as well as to the wider travel behaviour community. David has published over 600 papers in leading international transport and economics journals, as well as 16 books.

More from iMOVE CRC

Most Popular

ADVERTISEMENT

To Top