Air Pollution Monitoring and Prediction System

Air pollution is a silent killer, causing millions of deaths and preventable illnesses worldwide each year. It is like a "pandemic in slow motion," plaguing our health with diseases such as stroke, heart disease, lung cancer, and acute respiratory infections (WHO, 2021). In the UK alone, it leads to 36,000 premature deaths annually, costing £20 billion (Government's Committee on Medical Effects of Air-Pollutants COMEAP).

Although the pandemic-induced lockdowns resulted in a significant drop in global emissions in 2020, the easing of lockdown restrictions has led to a surge in pollution levels surpassing pre-pandemic levels. Despite various actions taken by governments, including the establishment of clean-air-zones, poor air quality is projected to persist until 2050 (OECD, 2019).

Scientists recommend adopting a dodging approach to protect vulnerable individuals, such as those with respiratory illnesses like asthma and bronchitis. These individuals are at a higher risk of developing complications or even dying due to exposure to high pollution levels (European Public Health Alliance, 2020). However, existing solutions that provide city-wide information are ineffective for a dodging approach, as vulnerable individuals struggle to determine specific locations within a city or town to use or avoid during walks, journeys, or exercise. To address this issue, a solution providing pollution data at a much smaller scale, such as the postcode-units level, is required. However, monitoring equipment for each of the approximately 1.7 million postcode-units in the UK is impractical. Nevertheless, the thousands of operational emission sensors across the UK (BBC, 2019) can provide enough data to develop models for all postcode-units with the right tools.

This project aims to develop a system that can provide pollution data to users using machine learning algorithms, GIS data, telematics, weather data, and big data analytics. The system will be accessible through a web and mobile app and will include the following components:

V1. LiveTap

V1 of the project focuses on LiveTap, an initial version that aims to provide real-time pollution data to users.

V2. Air-PoT

V2 of the project, led by a consortium of experienced experts in business, technology, and air pollution science, aims to leverage the advanced UK infrastructure of thousands of IoT sensors via a centralized and harmonized platform. This platform will be utilized by both sufferers and data-led organizations.

V3. PASS

V3 of the project aims to develop a system that provides postcode-units-specific pollution data to users. It will utilize machine learning algorithms, GIS data, telematics, weather data, and big data analytics. The system will offer a web and mobile app interface with live, future, and city/town analytics dashboards.

To learn more about our work, you can refer to the following publications:

2023:

Egwim, C.N., Alaka, H., Pan, Y., Balogun, H., Ajayi, S., Hye, A. and Egunjobi, O.O. (2023), "Ensemble of ensembles for fine particulate matter pollution prediction using big data analytics and IoT emission sensors", Journal of Engineering, Design and Technology, Vol. ahead-of-print No. ahead-of-print.

2022:

Sulaimon, I.A., Alaka, H., Olu-Ajayi, R., Ahmad, M., Ajayi, S. and Hye, A. (2022), "Effect of traffic data set on various machine-learning algorithms when forecasting air quality", Journal of Engineering, Design and Technology, Vol. ahead-of-print No. ahead-of-print.

2021:

Balogun, H., Alaka, H. and Egwim, C.N. (2021), "Boruta-grid-search least square support vector machine for NO2 pollution prediction using big data analytics and IoT emission sensors", Applied Computing and Informatics, Vol. ahead-of-print No. ahead-of-print.

Balogun, H., & Alaka, H. (2021). An Application Of Machine Learning With Boruta Feature Selection To Improve NO2 Pollution Prediction. In EDMIC 2021 CONFERENCE PROCEEDINGS ENVIRONMENTAL DESIGN & MANAGEMENT INTERNATIONAL CONFERENCE: Confluence of Theory and Practice in the Built Environment: Beyond Theory into Practice Obafemi Awolowo University, Ile-Ife.

Balogun, H., & Alaka, H. (2021). Random Forest Feature Selection for PM10 Pollution Concentration. In EDMIC 2021 CONFERENCE PROCEEDINGS ENVIRONMENTAL DESIGN & MANAGEMENT INTERNATIONAL CONFERENCE: Confluence of Theory and Practice in the Built Environment: Beyond Theory into Practice Obafemi Awolowo University, Ile-Ife.

Sulaimon, I., & Alaka, H. (2021). Air Pollution Prediction using Machine Learning – A Review. In EDMIC 2021 CONFERENCE PROCEEDINGS ENVIRONMENTAL DESIGN & MANAGEMENT INTERNATIONAL CONFERENCE: Confluence of Theory and Practice in the Built Environment: Beyond Theory into Practice Obafemi Awolowo University, Ile-Ife

Other works are currently under review. If you are interested in contributing or have any inquiries, please send an email to dtilabuh@gmail.com.