Modelling on the Move Symposium – a write-up

This article describes a symposium hosted by the University of Oxford’s Transport  Studies Unit (TSU) on December 7th 2012. Entitled Modelling on the Move the event kicked-off a new series of seminars sponsored by the Economic and Social Research Council (ESRC). The aim is to explore how quantitative models can inform a transition to a low-carbon transport system, by “bringing together researchers and practitioners to discuss innovative ways of responding to pressing policy problems in transport”. The premise of the series is the interlinked problems of the obesity crisis, climate change and oil depletion.

My first thought, before the conference had even begun, was respect to the organisers for facing such overriding problems with our transport systems head-on, rather than fiddling around the edges or arguing over academic minutiae. This symposium aimed to tackle the ‘big issues’, against the grain of academia’s tendency to “tell us more and more about less and less” (Gallagher and Appenzeller, 1999).

Convener Rachel Aldred introduced the need for ‘systemic transition’ in the context of climate and energy objectives, obesity and economic crises and the rise of ‘big data’. It was certainly a broad remit. Below I sketch how researchers focussed on one or more of these objectives are harnessing new modelling techniques and collaborating to foresee the transport systems of the future.

Specifically there were talks on:

The slides and audio from these talks have been made available online, and can be accessed from the Modelling on the Move website by clicking on the above talk titles. Or, for a more general summary from one perspective, read on.

The first speaker introduced was Professor David Banister – Editor of a number of Transport Journals and head of the Transport Studies Unit (TSU) at the University of Oxford. David Banister also delivered the TGRG Hoyle lecture in 2010.

A major problem in transport studies was identified at the outset: that the solutions to transport problems must be long-term but focus in academia is usually short term. For example, 95% of transport energy use is from oil. Transport uses 60% of oil worldwide. If per capita carbon dioxide targets for 2050 are to be met, we must produce no more than 2 tCO2 each year, 0.75 from transport. This is less than ½ current consumption in the West, indicating the need for fundamental shift in behaviour or technology, or both. Modelling these shifts is a major challenge.

In general terms, three approaches for investigating the future are available:

  1. Forecasting – Projecting change based on extrapolating past trends using mathematical models. “American tradition” cautious, “no regrets” strategy. Time horizon ~10 yrs.
  1. Exploratory – prospective forecasting. Look at possible and plausible futures. Peter Schwartz has developed a 10-step structure for implementing this approach. Visions/scenarios of future. UK Foresight Project 2006 – looks at a range of different futures. At global level can create intersecting axes of the future.
  1. Visioning – backcasting. Longer term (20 -30 years) uses normative vision of future to identify the changes that would be needed in the near future to allow this to happen. About trend-breaking. Participatory methods combining quantitative and qualitative data.

David Banister advocated the backcasting approach, and highlighted its main components.

David Banister concluded his talk by emphasising that both long-term and short-term scenarios need to be included. These scenarios must be connected and self-reinforcing. “Some would say that we need to move away from consumption-based lifestyles” so future scenarios must incorporate the possibility for widespread shifts in ways of life. The social sciences agenda may be key to this as deterministic science alone cannot provide solutions to the current problems. In many cases solutions are not technological.

The next speaker was Saamah Abdallah – from the centre for well-being, nef. His interest lies in planning for well-being, and how this can inform transport planning decision making.

Much of his talk was informed by a report commissioned by Defra looking at examples of past shifts in behaviour. Saamah began by citing Cuba in the early 1990s, arguing that it illustrated how a radical shift in behaviour (and reduced energy consumption) can happen in a controlled manner, guided by the state.

James Woodcock provided a health perspective on the issue of transport modelling citing numerous statistics related to the health benefits and risks of different transport behaviours. Following the ‘big picture’ approach of the entire syposium, his talk portrayed health as a complex package that also included diet and mental health. Active modes of transport and especially cycling have huge potential benefits. However, the health benefits tend to accrue disproportionately to the elderly, while generally it is young people who cycle. This paradox has policy implications, the detail of which remains to be seen.

James presented some visually striking and informative data on variations in health impacts of transport by age. In the Netherlands, for example, cars are the most risky form of transport (in terms of chances of death per vehicle kilometre) yet cars become a relatively safe mode for older people. Mopeds and bicycles, by contrast, become relatively more dangerous as age increases.

Finally there was a dichotomy highlighted: the promotion of active modes as a form of transport and active travel as a medicine. Also, areas with a high uptake of travel by active modes do not always have positive environmental outcomes. Switzerland, for example, has very high rates of walking and cycling, yet is a more energy intensive society, per capita, than the UK.

 Karen Lucas talked about the social impacts of transport, and the modelling that can inform us of these impacts. With all this talk of new technology and options, it’s interesting to think about what “bottom of the pile” are doing in terms of transport. Many people don’t have well-being now, before welfare impacts of de-carbonisation even kick in. How then, is well-being to be maintained as energy intensive options become less viable. “Social justice in transport” is a very broad subject area, she said.

Transport disadvantage and social disadvantage are separate concepts. Their overlap can be considered as ‘transport poverty’. This can lead to inaccessibility and subsequently social exclusion which has knock-on impacts including the long-term effects of unemployment.

Methods-wise, qualitative approaches have been very effective at finding out identifying the experiences of the transport poor. Quantitative approaches are up-and-coming with many different foci. Empirical studies of social equity and transport focused solely on age or gender – ignores social social disadvantage. GIS studies on the other hand tended to create place-based measures of transport disadvantage, and ignore individual-level variability.

However, models have yet to properly address issues of social disadvantage in transport opportunities. “Transport studies people are so thick when it comes to social disadvantage”.

“Can we have people based measures of accessibility?” Karen pleaded: accessiblity for young people is totally different than accessibility for the elderly.

Professor Peter Jones talked about the evolution of modelling in transport studies

Peter’s talk began by explaining that each discipline develops around a paradigm – a way of framing and deciding on research questions. The paradigm is central to explaining the current direction of transport research.

There have been a number of “conceptual enlargements” in the transport studies paradigm, beginning with how to move vehicles around moving towards mobility of individuals:

  1. Moving cars – focussed on accommodating the needs of motor vehicles – realisation that the cars are actually designed to move people.
  2. Moving people. Efficiency of transporting people, tolls inversely proportional to number of occupants.
  3. Focus on activities – what are people achieving in their mobility: it’s about accessing goods and services, in some cases by travelling less than they did before. Problem: peoples’ need for accessibility is subjective – difficult to model.
  4. Attitudes – Recognition of the importance of beliefs, attitudes and social norms. Social issues and inequalities. But what about individual-level variability?
  5. Dynamics – incorporating the time-scales involved in shifts in transport behaviour. Policies take time to implement – must acknowledge this.

The overall message from Peter’s talk was of the need for a paradigm shift in transport studies, to account for the dynamic nature of behaviour, and to inform a paradigm shift in people’s behaviour.

Professor Mike Batty discussed new transport modelling methods.

Picking up where Peter Jones left off, Mike Batty described one development that is forcing a kind of paradigm shift in transport research: Big Data.

Oyster card data is a good example – over 1 billion records – which changes types of people who can even get their minds around this data. Spatial and a temporal scales have shrunk massively. In some senses it is our ability to understand the data that’s important. Talk was held together by a selection of case studies of modelling based on the new types of data that have been made available in the 21st Century, including flows of bikes, Oyster cards and flow data.

A video of tube usage over a week was used to drive home the point. Complex systems understanding is a much more organic bottom-up approach to modelling. Challenge is to make sense of the flows – recently made much easier due to new data sources.

Despite being impressed by the sheer size of the datasets analysed by Batty and his team at the CASA, I was left wondering how all this related to the need for systemic transition. Many of the methods seemed more descriptive than the normative models recommended by David Banister.


This was a truly stimulating and engaging event. Despite the ambition and breadth of the symposium, it held together surprisingly well and (dare I say it) helped to set the agenda for the transport modellers of the future. Whether the series will meet that aim (hope this article has not jinxed it) will can only been seen in the future, which can be glimpsed through Modelling the Futures website:

One thought on “Modelling on the Move Symposium – a write-up”

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s