A Leeds Beckett student has created an artificial intelligence (AI) tool that can predict damp and mould in homes.
Gulala Aziz, a PhD student at the university’s sustainability institute, is developing an AI tool to support housing providers in managing damp and mould in properties.
Gulala’s tool is designed to predict which homes will be more likely to develop mould, meaning homes which require may require more frequent inspections can be flagged to prevent risk to tenants, Housing Digital reports.
Created using “machine learning with explainable AI”, the tool has been trained using over 2,000 home inspection reports across 125 local authorities in England.
After the death of two-year-old Awaab Ishak from a respiratory condition caused by prolonged exposure to damp and mould in his social housing flat, Awaab’s law was introduced. It forces social landlords to investigate and fix dangerous damp and mould in set time periods, and repair all emergency hazards within 24 hours.
ne of 204,000 annual deaths in the UK linked to poor air quality in the home. The risk increases among vulnerable groups, causing up to 15% of childhood asthma and accounting for 1.7m lost school days every year.
Awaab’s death was one of 204,000 annual deaths linked to poor air quality in homes, with an increasing risk of death amongst vulnerable groups.
Gulala’s tool assesses a properties features to evaluate its risk, including wall insulation, heating costs and energy efficiency.
After three years in development, the AI model’s checks have been finalised. This work represents the first peer-reviewed application of AI on damp and mould risk, and is featured in a Nature Scientific report.
Speaking about the research that went into development, Gulala said: “Housing associations and local authorities are subject to increasing requirements and scrutiny – and this research can help then to take swifter, more effective action.”
Adam Hardy, senior research fellow at the Leeds Sustainability Institute, added: “This research represents a huge step forward in understanding our housing stock.”
Featured image via Canva
