PhD Thesis – in progress (2017 – 2021)
Participatory mobility dashboards
This research develops software platforms containing urban dashboard prototypes (which may be multi-platform web-based, a smartphone application, mixed reality or physical interventions) which primarily investigate Sydney (with opportunities for partnerships, projects and collaboration with several other cities worldwide).
Each prototype’s effectiveness is evaluated by their capacity to engage and empower citizen users. Several use cases are built and evaluated using this platform.
The use cases specifically address the issue of urban transportation/mobility tasks at both city-wide strategic planning and day-to-day performance-based levels.
Use Case 1 – Operational Transport Planning
This use case examines how rapid, short-term participation may be achieved in operational planning in the context of data being generated in Sydney, Australia.
- Short-term (real-time) analysis ; users
- Created a real-time dashboard of transport-related Tweets
- Each Tweet is processed using Stream Analytics, sentiment is calculated using machine learning through IBM Watson Tone Analyser
- These are presented on a ‘dashboard’ that shows current and historic sentiment values of the city.
- Such initiatives can be useful for detecting disruptions, events and general mood of citizens towards certain key words (services, projects, modes, politicians..).
- Long-term (12-month) analysis ; users
- Long-term collection of Tweets related to transport over past 12 months
- Natural language processing techniques applied (Topic, Tone & Sentiment)
- Detailed survey of 100+ Twitter users whose data collected
- Comparison between survey techniques and automated techniques – are these methods suitable?
- Automated approaches useful ; however mixture of methods identified significantly more robust approach.
- Publication under review
- Social Media as passive geo-participation in transportation planning – how effective are topic modelling & sentiment analysis in comparison with citizen surveys?
- Long-term (12-month) analysis ; system
- Over 800 million GPS traces and performance metrics collected over 12 months from entire NSW public transport network
- Series of interactive visualisations created to interrogate historic performance
- Linked with user data in analysis of Tweets as a proxy for network performance
- Publication under review
- The visual analytics of big, open public transport data – a framework and pipeline for monitoring system performance in Greater Sydney
Use Case 2 – Strategic Transport Planning
This use case examines how complex, multi-stakeholder, long-term participation may be achieved in strategic transport and land-use planning in the context of data being generated in Sydney, Australia.
Part one investigated the usefulness of transport accessibility planning support systems. This work is described in this chapter published in Handbook of Planning Support Science: “The Making of a Mega-region: Evaluating and Proposing Long-term Transport Planning Strategies with Open-source Data and Transport Accessibility Tools”
Part two involved the co-design of an urban planning instrument which integrates machine learning into strategic initiatives. Over 50 participants were involved in this study which involved using a variety of existing and custom-made digital tools towards planning a significant growth area in Western Sydney.
- Publication under review
- Towards the collaborative development and use of Machine Learning techniques in Planning Support Systems – a Sydney example
Use Case 3 – Local Transport Planning
The final use case will integrate learnings from the Use Cases above, applied to developing effective data-driven participation systems in a local cycling context.
Australian Population Explorer
Infrastructure Australia – Future Cities Research (2017 – 18)
At Arup I was part of the strategic transport modelling team, developed an accessibility model based on OTP and pushed ArcGIS Online Story Maps to their limits for our Infrastructure Australia client. The results of our work was the below report:
Future Cities: Planning for our growing population is the fifth paper in Infrastructure Australia’s Reform Series. It provides advice to Australian governments on improving the productivity and liveability of our largest cities as they grow over the next 30 years.
The paper models long-term growth scenarios for Melbourne and Sydney and assesses their performance across a range of indicators. These include performance of the transport network, access to jobs, environmental performance of the road network, access to and demand for social infrastructure, and access to and demand for green space.
Interactive Map (Explainer)
Accessibility modelling & the 30 minute city (2016 – 17)
Using open-source data and transit network tools to improve our understanding of accessibility – a case study of the Sydney ‘30-minute city’
10th International Symposium on Digital Earth & Locate17, Sydney – 2017
Planning Institute of Australia National Congress, Sydney – 2017
Long-term strategic transport modelling, visualisation and network analysis for all major Australian cities 2014-16
Curating interactive results of transport model results across Melbourne, Sydney, Brisbane, Adelaide and Perth as part of a series of long-term forecasting reports for the Department of Infrastructure and Regional Development. Streamlined outputs into interactive Google Earth (3D) and CartoDB outputs.
Melbourne, where are we going? GovHack Melbourne Entry 2015
‘Melbourne, where are we going?’ is an interactive web visualisation platform that allows the user to explore the relationship between both transport and housing in Victoria.
This won several awards at GovHack 2015 – including the Victorian State Government Data Challenge Award, Best Use of Interdisciplinary Data Award and was runner-up for the National Best Policy Insights Hack.
‘Big’ transport data analysis & visualisation of San Francisco UCL Master’s dissertation, 2014
Thakuriah, P., Tilahun, N., and Zellner, M. (2016) Seeing Cities Through Big Data: Research Methods and Applications in Urban Informatics.Springer: New York. (In Press)
A Big Data Mashing Tool for Measuring Transit System Performance
Gregory Erhardt, Oliver Lock, Elsa Arcaute and Michael Batty, University College London, UK.
Presented at the Big Data and Urban Informatics Conference, Chicago 2014.
Keeping track – the fusion of large, automatically-collected transport data in capturing long-term system change
Oliver Lock, Gregory Erhardt
Presented at the Australian Institute of Traffic/Transport Planning and Management Conference, Brisbane 2015.
City of Culture – Interactive 3D Map of London and cultural data using three.js – UCL Master’s major assignment
Lightweight, virtual, 3D browser-based model of London overlayed with cultural data mined from Wikipedia. Also explored the concept of embedding VR technologies into the browser (Oculus Rift).
City of Culture – Visualisation of V&A Museum artefacts in D3.js – UCL Master’s major assignment
Interactive visualisation of the global and temporal spread of artefacts in the V&A museum, covering ancient, postclassical, early modern, mid-modern and contemporary pieces.
A multivariate UK Census data-mashing tool using R – Gentrification in London example – UCL Master’s major assignment