Regional Science Academy, in cooperation with NECTAR Cluster 6, promoted a Special Academic Session ‘Big Data, Transport Policies and Accessibility’, organized by Pierre Zembri Karst Geurs Aura Reggiani Peter Nijkamp and Hans Westlund. The event was held in Champs-sur-Marne (France), the 2nd and 3rd of July 2016.
This session promoted a thorough and academic reflection on the nature and implications of big data, searching for linkage between spatial accessibility issues and big data needs and exploring the implications of spatial big data systems for transport management and policies, with the following programme:
1. Opening session:
Peter Nijkamp (Tinbergen Institute Amsterdam)
NECTAR meeting the Regional Science Academy
2. The Voice of Regional Science
Michael Batty (UCL, London)
Big Data, Transport and Accessibility
Folke Snickars (KTH, Stockholm)
Big Data Strategy from a University perspective
Tomas Dentinho (Univ. Açores)
Data Requirements for Information Decision Systems of City Managers
3. Great Minds in Regional Science
Philippe Poinsot (LVMT, Marne-la-Vallée)
4. Round Table Discussion
Olivier Bonin, Mike Batty, Tomaz Dentinho, Karst Geurs, John Osth, Peter Nijkamp, Aura Reggiani, Folke Snickars, Pierre Zembri
After this session, several communications from members of the NECTAR Cluster 6 were presented and the event was concluded with a meeting between Regional Science Academy and NECTAR Cluster 6, aiming at planning further collaborations.
Programme and Summary
This combined meeting of the NECTAR cluster on ‘Accessibility’ and members of the Regional Science Academy was a logical follow-up of an ABC that took place earlier this year in Stockholm on ‘big data’. In the Paris meeting the focus was not so much on ‘big data’ per se, but on the challenges and opportunities of using ‘big data’ in transport systems, with particular emphasis as accessibility problems.
The meeting had a mixed character: the focus was partly on empirical research and big data treatment, and partly on a thorough reflection and conceptualisation of ‘spatial big data’. The framing of these issues was introduced by a comprehensive review of the issues concerned by Michael Batty, who in the Voice of Regional Science series convincingly demonstrated the potential of these new approaches by means of an empirical illustration for the London transport system and the use of the Oyster Card. Also the presentation by Karst Geurts highlighted the importance of ‘spatial big data’ issues in transportation research; followed by a presentation by Tomaz Dentinho. A welcome addition to the debate was also offered by Philippe Poinsot, who in the ‘Great Minds’ session provided a very informative presentation of the work of the great French scientist Jules Dupuit, who laid the foundations for a social economic analysis of transport investments.
‘Big data’ has indeed become an important research and policy issue in the past years, not only in the ‘hard’ sciences, but also in the social sciences and humanities. Research on ‘spatial big data’ has rapidly followed the great many challenges in ‘big data’ systems. Clearly, it is still an open question whether big data provide also better insights; this depends on the framing of the research question, on the appropriate indicators for testing meaningful propositions, and on the smart selection of theoretically founded research issues.
Clearly, ‘big data’ does not only mean large volumes of data. In the latter case, multivariate statistics might also be a valid option. But ‘big data’ also refers to complex dynamic systems, to non-linear evolutionary data patterns, and to multi-actor interdependencies. First attempts to deal with such challenging research tasks in the social sciences can be found in computational neural networks (CNN) and social network analysis (SNA). But in current ‘big data’ systems we have to link sometimes a set of ‘big data clouds’ to another set of ‘big data clouds’ , and this is a formidable research challenge. Examples in our digital world can be found in GSM systems, GPS systems, sensors, camera data, digital network information, inter-active social media (Facebook, Twitter, etc.).
A clear concern of the use of ‘big data’ in transportation research is formed by the consistent definition and statistical treatment of massive volumes of ‘big data’. This also prompts questions on consistent data comparison and on open access to ‘big data’ (e.g., mobile phone data) under strict privacy conditions. It is foreseeable that data warehousing under open access conditions will become a major challenge to the future of data-driven research in transportation science.