UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Modelling individual accessibility using Bayesian networks: A capabilities approach

Bantis, Athanasios; (2020) Modelling individual accessibility using Bayesian networks: A capabilities approach. Doctoral thesis (Eng.D), UCL (University College London). Green open access

[thumbnail of Bantis_10115960_Thesis.pdf]
Preview
Text
Bantis_10115960_Thesis.pdf

Download (35MB) | Preview

Abstract

The ability of an individual to reach and engage with basic services such as healthcare, education and activities such as employment is a fundamental aspect of their wellbeing. Within transport studies, accessibility is considered to be a valuable concept that can be used to generate insights on issues related to social exclusion due to limited access to transport options. Recently, researchers have attempted to link accessibility with popular theories of social justice such as Amartya Sen's Capabilities Approach (CA). Such studies have set the theoretical foundations on the way accessibility can be expressed through the CA, however, attempts to operationalise this approach remain fragmented and predominantly qualitative in nature. The data landscape however, has changed over the last decade providing an unprecedented quantity of transport related data at an individual level. Mobility data from dfferent sources have the potential to contribute to the understanding of individual accessibility and its relation to phenomena such as social exclusion. At the same time, the unlabelled nature of such data present a considerable challenge, as a non-trivial step of inference is required if one is to deduce the transportation modes used and activities reached. This thesis develops a novel framework for accessibility modelling using the CA as theoretical foundation. Within the scope of this thesis, this is used to assess the levels of equality experienced by individuals belonging to different population groups and its link to transport related social exclusion. In the proposed approach, activities reached and transportation modes used are considered manifestations of individual hidden capabilities. A modelling framework using dynamic Bayesian networks is developed to quantify and assess the relationships and dynamics of the different components in fluencing the capabilities sets. The developed approach can also provide inferential capabilities for activity type and transportation mode detection, making it suitable for use with unlabelled mobility data such as Automatic Fare Collection Systems (AFC), mobile phone and social media. The usefulness of the proposed framework is demonstrated through three case studies. In the first case study, mobile phone data were used to explore the interaction of individuals with different public transportation modes. It was found that assumptions about individual mobility preferences derived from travel surveys may not always hold, providing evidence for the significance of personal characteristics to the choices of transportation modes. In the second case, the proposed framework is used for activity type inference, testing the limits of accuracy that can be achieved from unlabelled social media data. A combination of the previous case studies, the third case further defines a generative model which is used to develop the proposed capabilities approach to accessibility model. Using data from London's Automatic Fare Collection Systems (AFC) system, the elements of the capabilities set are explicitly de ned and linked with an individual's personal characteristics, external variables and functionings. The results are used to explore the link between social exclusion and transport disadvantage, revealing distinct patterns that can be attributed to different accessibility levels.

Type: Thesis (Doctoral)
Qualification: Eng.D
Title: Modelling individual accessibility using Bayesian networks: A capabilities approach
Event: UCL (University College London)
Open access status: An open access version is available from UCL Discovery
Language: English
Additional information: Copyright © The Author 2020. Original content in this thesis is licensed under the terms of the Creative Commons Attribution 4.0 International (CC BY 4.0) Licence (https://creativecommons.org/licenses/by/4.0/). Any third-party copyright material present remains the property of its respective owner(s) and is licensed under its existing terms. Access may initially be restricted at the author’s request.
UCL classification: UCL
UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Civil, Environ and Geomatic Eng
URI: https://discovery.ucl.ac.uk/id/eprint/10115960
Downloads since deposit
109Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item