Use of Artificial Intelligence to Evaluate Effectiveness of UK Clean Air Zones for Improving Air Quality and Health and Wellbeing

   Faculty of Science and Engineering

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Prof Dhaval Thakker, Dr Sheen Mclean Cabaneros, Prof N Mishra  Applications accepted all year round  Self-Funded PhD Students Only

About the Project

Air pollution causes up to 36000 deaths in the UK and over 6.5 million deaths globally. World Health Organisation (WHO) has recognised that Air pollution is the most significant environmental health risk we face today. The UK has been referred to the European Court of Justice for failing to take enough action to prevent breaches of NOx pollution limits. As a result of the general awareness of the pollution risks, and legal requirements, the UK government has compelled several urban areas around the country that breached EU (European Union) regulated limits to bring in drastic interventions to reduce air pollution. Thirty-three local authorities with exceptionally high pollution levels are required to implement air quality plans, of which Clean Air Zones (CAZ) are a key component. There are four types of clean air zones, Class A to D, with various restrictions on types of business and personal vehicles. This has come at a high cost to the public money, with the focus being to reduce pollution and improve health and wellbeing. However, evidence about whether CAZ improve air quality and health or wellbeing is lacking. 

This PhD will utilise Artificial Intelligence techniques to devise a technical framework to measure the effectiveness of the CAZ implementations in terms of reducing Air pollution, improving health and wellbeing. State-of-the-art deep learning models will be investigated using the data collected within and outside the scope of the CAZs to create spatial maps of NOx and PM concentration levels. The spatial analysis will assess the models of different CAZs, and their impact on the areas within the CAZ compared to areas with a variety of distances from the CAZ. For instance, data from monitoring sites outside the CAZs will be used during the spatial interpolation scheme as proxies for unmonitored areas to provide data for comparative analysis. As a final step, several stochastic techniques, such as Bayesian regularisation and Monte Carlo simulation, shall be investigated during the model development stage to quantify the uncertainty of the results.

We will carry out this PhD research focusing on three CAZs- Bradford (North), Birmingham (Midlands) and Portsmouth (South) where the data is available to support the research, and an advisory external panel members from these local authorities will be formed to be part of the project advisory team.

Eligibility and entry requirements

Applicants should have a minimum 2:1 degree in Computer Science or related subject. A taught MSc or Masters by Research in a relevant subject or relevant laboratory experience would be an advantage.

How to apply IMPORTANT – Please include the project title and proposed supervisor in your application

Computer Science (8) Environmental Sciences (13) Geography (17) Mathematics (25)

Funding Notes

Self-Funded PhD Students Only.
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.