Transportation network modelling and simulation under disaster scenarios

Project description

Road networks often get disrupted due to natural disasters (floods, earthquakes, bushfires, etc.), traffic incidents, system failures, and special events. In India, floods are a major problem, especially in the monsoon, whereas, in Australia, bushfires are a major problem during the summer. Understanding road transportation networks, resilience to these disruptions is essential to maintaining their continuity and ensuring efficient and effective movement of goods and people. This project aims to use crowdsourced data and simulation to evaluate and compare the resilience of Indian and Australian cities.
The road network resilience shall be assessed using two approaches:
i) Data-based: Indicators such as speed and travel time shall be collected using crowdsourced data providers to assess whether a network can maintain a certain level of performance during disruptions. Data shall be collected on typical days (no major disaster), and disaster scenarios (pre, mid and post) for at least 10 Indian and Australian cities. The key questions answered through this approach are:
a. Which city (or parts of the city) is more resilient in terms of day-to-day congestion? Which city has slower congestion propagation, and which city has faster dissipation?
b. What makes a city more resilient? What is the role of road network structure, land use and socio-demographic characteristics?
c. What are the impacts of past and ongoing disasters?
ii) Simulation-based: The project involves creating hypothetical disruption scenarios and measuring the impacts in various Indian and Australian cities through macro simulation exercises. The demand data needed for the simulation shall be estimated using machine-learning techniques. The developed models shall be calibrated and validated using field and crowdsourced data. The key questions answered through this approach are:
a. Given the same scale of the disaster, how do different cities perform?
b. What is the critical failure limit for each network?

Outcomes

The study will contribute to understanding the resilience of road transportation networks by identifying the key factors that affect their performance and functionality during and after disruptions. The developed models will help the end users, such as planners or government organisations, for traffic management, disaster simulation, and disaster response strategies. These strategies involve improving road infrastructure and connectivity, implementing new traffic management systems, or updating emergency response protocols.
Specific deliverables include scripts to extract network and travel time data; an executable/dll file to run the macro simulation scenarios; user documentation with formats for input and output that would be determined through usability testing; sample input and output files; a final report.
The research will result in at least 3 research articles published in high-quality journals such as Nature Sustainability, International Journal of Disaster Risk Reduction, and Progress in Disaster Science. Also, the work will be disseminated at national and international conferences.

Essential capabilities

Familarity with programing using Python, background in machine learning and statistics

Desireable capabilities

Comfortable with the fundamentals of transport network modelling and simulation

Expected qualifications (Course/Degrees etc.)

Masters in related fields and research experience.