PhD Thesis: Understanding cycling behaviour through visual analysis of a large-scale observational dataset
Supervised by Professor Jo Wood (giCentre, City University London). Examined by Professor John Parkin (UWE), Professor David Bawden (City University London)
The emergence of third-generation, technology-based public bikeshare schemes offers new opportunities for researching cycling behaviour. In this study, data from one such scheme, the London Cycle Hire Scheme (LCHS), are analysed. Algorithms are developed for summarising and labelling cyclists’ usage behaviours and tailored visual analysis applications are designed for exploring their spatiotemporal context.
Many of the research findings provide support to existing literature, particularly around gendered cycling behaviour. As well as making more discretionary journeys, women appear to preferentially select parts of London associated with greater levels of safety; and this is true even after controlling for geodemographic differences and levels of LCHS cycling experience. One hypothesis is that these differences represent diverging attitudes and perceptions. After developing a technique for identifying cyclists’ workplaces, these differences might also be explained by where cyclists need to travel for work and other facilities. An additional explanation is later offered that relates to the nature of cyclists’ estimated routes. The size and precision of the LCHS dataset allows under-explored aspects of behaviour to be investigated. Group cycling events – instances where two or more cyclists make journeys together in space and time – are labelled and analysed on a large scale. For certain types of cyclist, group cycling appears to encourage more extensive spatiotemporal cycling behaviour and there is some evidence to suggest that group cycling may help initiate scheme usage.
The domain-specific findings, emerging research questions and also behavioural classifications are this study’s principal and unique contribution. A second contribution relates to the analysis approach. This is a data-driven study that takes a large dataset, measuring use of a relatively new cycle facility, and uses it to engage with research questions that are typically answered with very different datasets. There is some uncertainty around how discriminating and generalisable LCHS cycle behaviours may be and which variables, either directly measured or derived, might delineate those behaviours. Visual analysis techniques are shown to be effective in this more speculative research context: numerous behaviours are very quickly explored and understood. These techniques also enable a set of colleagues with relatively limited analysis experience, but substantial domain knowledge, to participate in the analysis and a general argument is made for their use in other, interdisciplinary analysis contexts.
Chapter 1 introduces the research context and objectives: to investigate whether and how the LCHS dataset can be used to contribute to, and extend, existing research in cycling behaviour. The primary and secondary contributions of the PhD thesis are also stated.
In Chapter 2, some of the challenges associated with using, or repurposing, the LCHS dataset to study cycling behaviour are repeated, along with a critical discussion of visual analysis techniques and a justification for their use in this research.
Chapter 3 describes the LCHS dataset in detail: how data are recorded and also external information used to supplement the customer database. A number of behavioural variables for summarising individual cyclists’ scheme usage are created and the main visual analysis application for exploring the LCHS dataset is introduced. Decisions around these behavioural variables, and design decisions for the visual analysis application, are justified with recourse to literature within Transport Studies and Information Visualization.
Chapter 4 is the first substantive findings section. The current literature on gender and urban cycle behaviour is large and the chapter focusses on differences in men’s and women’s cycle behaviours using the software described in Chapter 3. Substantial differences between men’s and women’s usage are found that might relate to differences in the types of men and women subscribing to the scheme, but also possibly to more fundamental differences in men’s and women’s approaches to cycling. The chapter concludes by reflecting on the level and detail of findings that were achieved through simply exploring the LCHS dataset.
Chapter 5: In Chapter 4, claims about apparent commuting behaviour are made by visually scanning spatiotemporal travel behaviours. As the analytical enquiry progresses, observed behaviours are studied more formally. In Chapter 5 a technique for automatically labelling commuting journeys is developed and commuting behaviours are investigated in some detail. The approach to labelling commuting journeys may be applied to other origin-destination transport datasets, but the chapter concludes by focussing on the implications of the workplace classification on this study and on earlier attempts at explaining bikeshare cycling behaviours.
The LCHS dataset provides a total, population-level record of customers’ scheme usage. Chapter 6 demonstrates how this fact allows a new, previously under-researched aspect of cycling behaviour to be studied: that of ‘group cycling’. Group cycling is defined as journeys made between at least a pair of bikeshare cyclists together in space and time. Once labelled, group-cycling journeys are explored using the software introduced in Chapter 3. A new application is also created here that allows individual group-cycling networks to be explored over space and time. The chapter concludes by reflecting on the findings and their implications for wider research in Transport Studies.
Chapter 7: Many findings from this analysis relate to interesting and distinct spatial cycle behaviours. A limitation here is that with only the origins and destinations of cycle journeys, nothing is known about the nature of likely routes that might be encountered by cyclists. In Chapter 7, an attempt is made to address this issue by collecting information on estimated routes for every cycled OD pair in the dataset. From this information, heuristics on the nature of routed journeys are collected. A problem with this approach is that there is no means of knowing how closely estimated routes relate to customers’ actually cycled routes. The chapter therefore focusses on an aspect about which there is greater certainty: the bridge that is suggested by the routing algorithm for journeys that involve a river crossing. More detailed explanations for observed spatial travel behaviours are offered as a result of this analysis, along with a discussion around the extent to which these explanations might be formally described and quantified.
Chapter 8: The Conclusion chapter returns to the three objectives introduced in Section 1.2. The study’s domain-specific contributions are outlined, along with an argument for the analysis approach. Some time is spent reflecting on the implications for academic research in Transport Studies, applied Information Visualization and the data-driven social sciences, but also for those working in public policy who wish to promote urban cycling and those in operations responsible for the running of bikeshare schemes. By critically engaging with the limits of this research as well as its contributions, an immediate research agenda is also specified. The chapter concludes by stating the thesis argued from this data analysis study.
Full record on City Research Online:
Beecham, R. (2014) Understanding cycling behaviour through visual analysis of a large-scale observational dataset. Unpublished Doctoral Thesis, City University London.
Publications as part of the thesis:
Wood, J., Beecham, R. & Dykes, J. (2014) Moving beyond sequential design: Reflections on a rich multi-channel approach to data visualization. IEEE Transactions on Visualization and Computer Graphics, 20(12), pp.2171-2180
Beecham, R. & Wood, J. (2014) Characterising group-cycling journeys using interactive graphics. Transportation Research Part C: Emerging Technologies, 47, pp.194-206.
Beecham, R., Wood, J. & Bowerman, A. (2014) Studying commuting behaviours using collaborative visual analytics. Computers, Environment and Urban Systems, 27, pp. 5-15.
Beecham, R. & Wood, J. (2014) Exploring gendered cycling behaviours within a large-scale behavioural dataset. Transportation Planning and Technology, 37(1), pp. 83-97.