Undertaking repairs to maintain housing stock and improve tenant wellbeing has become one of the greatest challenges facing housing associations today, particularly in an environment where the rent increase cap is 7% but costs of labour and materials have increased by 10-15%. Property condition remains the biggest single reason for referrals to the Housing Ombudsman, which indicates high levels of tenant dissatisfaction with repairs and maintenance services. The Ombudsman’s conclusion is that “more action is required to improve the quality of homes and service.”
With the government’s "Make Things Right" campaign and the recent introduction of Tenancy Satisfaction Measures (TSM) by the Regulator of Social Housing (RSH), further pressure is being put on housing associations to effectively track and respond to repairs, safety checks, and complaints, and provide secure and well-maintained housing.
On top of this, in the policy paper “Reshaping consumer regulation: Our implementation plan”, the RSH has stated that an “effective repairs and maintenance service” has a “substantial influence on the trust and confidence” which tenants have in their landlord and, therefore, are “central to delivering good quality homes”. These maintenance issues, along with matters related to transparency between housing providers and tenants, and energy efficiency within the sector, have driven the regulatory body to promise “a step change in the regulation of social housing” from April 2024.
So, it has never been more important for housing providers to rethink their repairs and maintenance processes, and leverage technology where possible to improve processes, cut costs, and most importantly, provide tenants with safe and well-maintained housing. The challenge of doing so can often seem daunting.
Could AI provide the solution?
Much has been written recently about Artificial Intelligence (AI), not all of it complimentary. But in the field of repairs and maintenance, the advent of AI-based technologies has brought a raft of new tools to support operations and aid in the planning and delivering of maintenance programmes, including smart sensors, chatbots, and virtual assistants.
One example of this is repairs scheduling – a notoriously difficult problem that is almost impossible to solve optimally with many associations relying on a combination of experienced personnel and rudimentary tactical tools, e.g. postcode mapping to work out daily routes. AI, however, can give landlords the ability to explore the vast expanse of possible scheduling options and find those that will minimise costs in a fraction of the time it would take a manual system. This includes incorporating geographical considerations into the solution, such as repairing a suite of roofs on the same site at the same time, rather than fixing them one by one in different locations.
Another application is AI-driven prediction, which uses AI-powered algorithms to predict both future customer behaviour and operational outcomes. The algorithms analyse the copious amounts of data that housing associations hold about their properties and their tenants in order to identify subtle patterns that indicate how outcomes are likely to play out.
By using this form of AI, housing providers could identify maintenance issues, such as boiler breakdowns, mould growth risk and roof repair, at a much earlier stage and gain a forward-looking view of their scheduling process. When combined with information from smart devices in people’s homes that monitor things like temperature and humidity, pre-emptive maintenance becomes a very real possibility.
This has the power to not only allow organisations to quickly take appropriate action but would also minimise the need for tenants to report issues, or complaints, as the matter could be effectively resolved with minimal disruption to the tenant. According to the English Housing Survey, in the 2020-21 period, nearly 33% of social renters considered filing a complaint, but 27% of them decided not to take any action as they believed that it would not yield any response. By making the most out of new technology available to them, housing providers can identify where problems are going to appear before they have manifested, saving costs and time for the housing provider and instilling confidence and reassurance in the tenant.
How can Housing Associations take their first steps towards AI predictive maintenance?
First, the organisation needs to have good quality data. The richer the data and historical information available, the better the predictions will be.
Secondly, the organisation needs a clear vision of what it wants to predict and how it will use those predictions. This is arguably more important than the technical side, so engaging with stakeholders, including senior leaders, decision-makers, and those who will be acting on the results, from the outset is critical. This ensures that the predictive insights align with business objectives and deliver real operational and commercial benefits for the organisation, and safe, secure and well-maintained housing for the tenants.
Jaywing's housing maintenance solution
Supporting England’s largest specialist housing and care provider