Sam Dugmore, systems development and support manager at Wolverhampton Homes, explains how AI has helped Woverhampton Homes to spot properties at greatest risk of damp and mould to help keep residents safe
One of the most persistent challenges facing housing providers today is responding quickly to issues such as damp and mould. These aren’t new problems for social landlords, but expectations are changing around how and when they are addressed.
Awaab’s Law has led to new guidance from the Housing Ombudsman, which means housing teams must be more responsive to residents’ concerns about the state of their homes. This includes the directive to reduce damp and mould risk to better protect people’s health.
These new developments could see tools such as artificial intelligence (AI) start to prove their worth in the housing sector. By helping housing providers to understand where the risk of damp and mould is highest, AI can support smarter decisions on where, when and how to act to deliver safer homes, faster repairs and better outcomes for residents.
At Wolverhampton Homes, we’ve been exploring how AI can be used to help us anticipate issues with damp and mould before they emerge. This has been a game changer for us, as AI has allowed us to predict damp and mould with up to 98 per cent accuracy in about two-thirds of the 21,000 homes we used to test the technology.
Since then, we’ve improved the way we manage damp and mould risk and gained some great insights into what’s causing these problems in the first place. This has helped us reshape the way we plan and deliver our housing services.
A smarter way to spot the signs
As the demand for social housing keeps growing, the traditional ways of identifying damp and mould such as regular inspections or monitoring residents’ complaints, won’t be enough to help landlords stay one step ahead of developing problems. We are working to prevent issues in the first place and AI has been a key part of this.
In a pilot project with NEC Housing, we’ve used AI technology to cross-reference data such as the age of housing stock ventilation levels and repair histories, with local geographical and weather information to build a risk index for damp and mould in our homes. For example, a 1960s flat in a humid area with historical damp and mould issues would likely be flagged as a high risk, even if no recent complaint has come in. This insight has allowed us to schedule inspections and maintenance in homes on the at-risk list before issues arise, which prevents more serious damage or significant repairs that are often disruptive for residents.
Faster responses, better data
We noticed that when the information going into the AI tool was out of date or incomplete, the accuracy of the predictions fell from 98 per cent to 70 per cent. To help close that gap, our housing officers and customer services colleagues now treat accurate record-keeping as a key part of their roles. If a housing officer spots condensation or learns that a resident has developed breathing issues while out on a visit, they can add this new information to our housing system straight away. This also updates the property’s risk score for damp and mould in real time, which can automatically move a home up the priority list.
Planning for the long term
Beyond just handling day-to-day repairs, AI will help us to make more informed, long-term investment decisions on our properties. By analysing data across all our properties, we could see patterns we might otherwise have missed such as certain types of homes or specific areas within the city that have persisting mould problems.
Using this model we have been able to predict (with a high degree of accuracy) and identify high priority cases within our stock. This data will in future support Wolverhampton Homes in managing our programmes of work and carrying out any required upgrades and retrofit to our properties, allowing us to look more closely at the design of homes as a result. Instead of relying on occasional surveys or anecdotal evidence from our housing officers, we now have a clearer view of where investment is needed backed by data.
Putting residents first
The real value of AI isn’t just in operational efficiency. It’s in the impact on residents’ lives. Residents are now getting their issues resolved faster, face fewer repeat problems, and experience less disruption. When homes are in good repair, they contribute to good health, especially for children, older adults, and people with respiratory conditions and other vulnerabilities. Plus, if we notice changes in residents’ behaviour, such as a resident that suddenly starts using less heating, we can reach out to see if they’re struggling with rising costs and may be able to offer some financial support if needed. The preventative measures we’ve been able to put in place have enabled us to play our part in protecting our residents’ health and wellbeing.
Making it work: challenges and lessons
The introduction of AI across the social housing sector may not be without its hurdles. Any data fed into the tool will need to be cleaned and kept up to date. Colleagues may have to adapt to new ways of working too, but our trial has shown that the outcomes can be worth it.
We’ve managed these challenges through a phased rollout of AI, colleague training and having individual AI champions in our teams to support the change.
While still in its infancy, our use of AI has already prompted us to look at how the technology can be used in other areas of housing, such as to monitor energy efficiency in residents’ homes, improve supply chain planning and cut carbon emissions with property retrofits.
A better way to build
As local authorities and housing providers continue to modernise, AI has the potential to drive lasting change, not just as an innovative piece of technology, but as a catalyst for successful digital transformation.
With increasing housing demands, tighter regulations and limited budgets, housing providers will continually need to do more with less.
By turning data into insight with AI, the sector can move away from reacting to problems and prevent them altogether.
