You Say You Want a Revolution

Some say there’s a revolution in transport. But to have a revolution, you need revolutionaries.

Transport is certainly undergoing a transformation in the digital age, forging ahead rapidly in the buzzwords of ‘big data’ generation and the development of ‘Internet of Things’ connections. Transport has become ‘smarter’, i.e. more automated, more dynamic, more shared, and more personalised, such that ‘mobility as a service’ (MAAS) is billed as the ‘mode’ of travel of the future. But is MAAS the transport revolution of the 21st century in the same way that the train or the automobile revolutionised transport in the 19th and 20th centuries by creating mass markets of revolutionaries willing to adopt new ways of getting around and even new lifestyles because of the new transport technology?

According to Professor Cristina Pronello of the Université Technologique de Compiègne and Politecnico di Torino, MAAS will only revolutionize transport and travel behaviour if it is developed in a ‘user-centric’, transparent and integrated manner. Which, she said, means MAAS should not be developed and imposed on society by the big corporate players such as Google, Amazon, Apple, etc.

In a talk she recently gave at the University of Greenwich, she compared the tech company travel solutions to an app which was developed for Turin, Italy as part of a European Commission-funded project she led. She called her app an urban navigator, not a journey planner, and pointed to the depth of real-time, multi-modal feeds and customization it offered to users to support their travel decisions.

This ticked the user-centric and integrated boxes, but the Professor admitted difficulty both in establishing data-sharing arrangements with the various transport providers to build truly real-time integration and also in recruiting participating travellers to ensure user-centricity.

I couldn’t help thinking that Google Maps has none of these problems. Most people have it pre-loaded on their smartphone, and the company set the original standard for open data feeds. And if it has not quite the depth and real-time reactions of the app she discussed, I have met Google employees and am given to understand that they are constantly improving the multi-modal integration and accuracy of their mapping – journey planning – navigation tools through validated historic trips which users have themselves tracked and reported.

So what about transparency? There’s the rub. The European project embarked on numerous contracts, enforced a standard data format, and created an open data portal. Anyone with the skills could see where the data was coming from and how it was driving the app’s results. The tech giants are much more opaque, and most would agree that they are more motivated by the bottom line and intellectual property rights, than by a public service mission.

Yet, as was discussed in another presentation in Greenwich, traditional methods and local governments can fail to address social equity in their transport provision as much as corporations do. Although statistical methods tend to be based on assumptions about behaviour in the pursuit of explaining why travel patterns occur and how societal trends may influence those patterns, they are rarely then used to influence decisions to create more equitable patterns. And the models themselves are often black boxes, with calculations undertaken within proprietary software.

In contrast, the algorithms of data analytics are based on no assumptions at all. They seek to learn patterns to accurately make predictions, not to explain how or why those futures have come to pass. If such patterns are most likely to create a viable, successful Mobility as a Service, then transport practitioners should surely be turning to algorithms instead of assumptions, and perhaps also to companies like Google for their expertise – and for access to all those potential revolutionaries already using Google Maps or paying for multi-modal transport on their smartphones.

And yet. Maybe the companies of the digital age should only be supporting the revolution, not leading it. If they are producing the algorithms and finding the patterns, there is still a place for transport planners, land use planners, and civic society to shape those patterns to be more equitable, more affordable, more sustainable. The beauty of those algorithms is that if they happen to find a pattern in the shape of a virtuous circle, they’ll advertise it and disseminate it without asking why, and that really would be a revolution.

Mobility vs Accessibility: new evidence for an old debate

I was at a public exhibition many years ago where I was approached by a rather aggressive environmental campaigner. He told me that if public sector transport planners like me really wanted to promote sustainable travel, then we’d all live and work within the same Local Authority area. Everything we did should be local and we shouldn’t really need to go anywhere, and then we wouldn’t be emitting all that carbon travelling. As I lived in another District from where I was working, albeit only 10-12 miles away, I naturally did no more than nod and smile politely.

Inside, I was thinking: Yes I agree that we need to reduce our transport emissions and impacts on the climate, but my husband is the one that lives next to his work and I had to find a job nearby. Yes I’d like a shorter, more convenient commute, but there wasn’t a job in my field, never mind at the level I was looking for, advertised within my District at the time. Yes I prefer to travel by sustainable modes, but I do take the train to get here, whereas I might have to drive to other jobs at a similar or shorter distance. Yes, but…!

Ok, enough of the protests in my head that clearly have been yearning to break free for far too long. My point in recalling this story is that the man’s superficially inane, impractical argument does have a grounding in a fundamental principle of transport that many transport planners, never mind transport users, often overlook. Transport planners tend to focus on creating and promoting options (read new infrastructure or services) for mobility, rather than accessibility.

Yet people travel for the purpose of accessing a job or a shop or a friend’s house, and travel further if those things they are trying to access are further away. The further they travel, the less mobility options they have, which may result in a poor choice between car-dependence and isolation. The latter I add as we consider the impact of new online technologies on accessibility over mobility. See a great blog on this by @alikirkbride for #LTTMobilityMatters.

Moreover, I have recently discovered that the concept that humans seek accessibility rather than mobility can be backed up scientifically. In the last decade, researchers [1-4] have used big data from mobile phone call records and social media to show that human movement follows certain patterns, namely:

  • Most people can be found in a few predictable places (home, work) most days of the week at the times (night and day) where you’d expect to find them there.
  • Most people make more short trips than long trips, and the distribution of short trips follows a certain pattern, decreasing with distance, up to a threshold.
  • At which point you have a different pattern where people who travel further can be found in expected places more often and have fewer irregular trips.
  • And, those people who travel further tend to live where there is less density – of population, employment, opportunities, activities – than those who travel shorter distances.

It is this last point that is key. None of the studies are looking at mode of travel, but they say something very basic about travel behaviour. Namely, that people are not choosing which trips to make to minimise journey times or distance travelled, even if that may influence modal choice. No, they are choosing which trips to make based on where the destinations are which they are trying to reach. They will choose the nearest destination that meets their need or desire or nearest ‘intervening opportunity’ as one study calls it [2].

Thus, transport planners should be as aware as land use planners of the importance of place-making, of mixed-use development, of walkable neighbourhoods. Discussing those is a whole other blog, so I’ll leave it there, but in a twisted way, that man who chastised me long ago had a point. If we could work at the sort of job we wanted, shop for whatever we needed, socialise with our friends and family and have our children in decent schools, all in the same area as our home, we would probably choose to do so. Then we would have more options for sustainable mobility (e.g. walking and cycling), which would be better for the environment and our health and make us more resilient to unforeseen events. And so transport planners would be planning for accessibility rather than mobility.

  1. Gonzalez, M.C.H., Cesar A. & Barabasi, Albert-Laszlo, Understanding individual human mobility patterns. Nature, 2008. 453(7196): p. 779-782.
  2. Noulas A, S.S., Lambiotte R, Pontil M, and Mascolo C, A Tale of Many Cities: Universal Patterns in Human Urban Mobility. PLoS ONE, 2012. 7(5): p. 1-10.
  3. Isaacman S, B.R., Caceres R, Kobourov SG, Martonosi M, Rowland, J and Varshavsky, A. Identifying Important Places in People’s Lives from Cellular Network Data. in 9th International Conference on Pervasive Computing (Pervasive). 2011.
  4. Song, C.Q., Zehui Qu; Blumm, Nicholas and Barabási, Albert-László, Limits of Predictability in Human Mobility. Science, 2010. 327: p. 1018-1021.