Using big data for intelligent urban planning

© hjl :: Flickr :: CC BY-NC 2.0

In the post-Snowden era, the collection and analysis of so-called 'big data' has increasingly become a highly controversial subject in public discussions around the world.

Proponents of privacy rights have convincingly demonstrated how the limitless gathering, storage and examination of virtually all kinds of bits and bytes not only by state actors, such as security agencies, but also by private commercial data miners all too often gravely violate citizens' basic rights.

Nevertheless, there are instances where the use of big data ultimately can be beneficial.

At the Beijing Institute of City Planning, Chinese urban planner Long Ying (who is also the founder of Beijing City Lab) and his colleagues are trying to find out what a certain type of traffic statistics might tell us about the link between patterns of spatio-temporal movement and socio-economic status.

To get around the sprawling Chinese capital, most travellers purchase smartcards as these are heavily subsidized and can be used for all sorts of public transport as well as for other services. Hopping on subway or bus lines, riders swipe their cards over readers, thus generating an immense data pool waiting to be analysed.

In a recent pioneering project, several Chinese researchers led by Long Ying studied the ubiquitous smartcard records of more than eight million passengers of the city's subway and bus systems from the years 2008 and 2010, respectively. In doing so, they were able to assess distinct travel patterns and to identify and characterize economically underprivileged residents.

Testing their hypothesis against the household travel survey (2010), a small-scale study (2012) and their own interviews with locals, the team discovered that those who frequently cover long distances tend to reside in faraway corners that are shunned by more affluent inhabitants of Beijing. In addition, erratic transit patterns might be an indication that the persons in question don't have steady employment or permanent housing.

Using both a traditional household survey and emerging new sources such as public transportation smartcard data, Long and his co-workers also developed a typology of four different groups of so-called 'extreme' commuters they dubbed 'early birds', 'night owls', 'tireless itinerants' and 'recurring itinerants'. Their tentative profiles reveal further details with regard to the close connection between specific travel patterns and social class.

Evidence from such big data surveys can be instrumental in revising public transit planning and design in general, and what is more, it can also help to devise a more equitable provision of social welfare or to direct resources for the construction of affordable housing, to name just two examples of social policy programmes.

New Scientist quotes Long as follows:

'Chinese people do not like to tell others their income. The government does not have a very effective way to know people's social economic status ... We hope our research can contribute to these projects.'

If you want to know more about survey methodology and findings, download the two relevant working papers (44 'Profiling underprivileged residents with mid-term public transit smartcard data of Beijing' / 57 'Early birds, night owls, and tireless/recurring itinerants: An exploratory analysis of extreme transit behaviors in Beijing, China') by Long Ying and his respective collaborators here.