In a rapidly evolving digital landscape, councils across the UK face an important challenge – the evolution of their public service delivery methods. Among these services, the allocation of social housing stands out and reflects a critical aspect of asset management for local councils as well as a crucial asset under local council purview. More people in the UK are struggling to afford or find a secure place to live than ever with over one million families waiting for social housing. In 2022 alone, 29,000 social homes were sold or demolished and less than 7,000 were built. The result of this being a daunting and extremely challenging task for local councils to get right.
Other than simply building enough homes for everyone, which presents itself as unlikely, there is no fix all solution. However, help may be closer than anticipated. Quietly humming in the undercurrent of the modern technological revolution, artificial intelligence (AI) and more specifically machine learning presents an opportunity to streamline and optimise the social housing allocation process and help councils manage their social housing to ensure allocation is as efficient as possible.
AI, machine learning and asset management
AI is a field in computer science dedicated to the creation of systems capable of performing tasks that would typically require human intelligence. Problem-solving, learning, perception and language understanding are some of the typical capabilities demonstrated by AI.
One crucial and very powerful subfield of AI is machine learning. Machine learning focuses on the development of computer algorithms and statistical models that enables machines to improve their performance in an area over time without being explicitly programmed to do so. Through refining their models as they are exposed to more data over time, machine learning becomes highly proficient in recognising patterns and trends that may be too complex for humans to manually identify.
Local councils across the UK oversee vast social housing stocks, providing affordable housing to millions of citizens. Social houses are also one of the most valuable and important assets that local councils possess and managing them effectively is of paramount importance. Until recently, asset management was more paper-work intensive than it was asset-intensive. However, as time has worn on and work levels and asset numbers have increased, it has been necessary to integrate asset management systems to optimise the task of managing assets.
Understanding asset aanagement
Asset management is the systematic process that entails the managing, monitoring and tracking of a company’s assets in a cost-effective manner. Assets may be tangible, such as social houses, or intangible such as software. Effective asset management serves as the backbone of any successful organisation, whether large or small. A robust asset management strategy empowers organisations to make well-informed decisions based on their assets such as when to replace assets nearing the end of their productive life-cycle, where to allocate assets to ensure they are being used optimally and when to schedule maintenance of assets to reduce unexpected costs.
How incorporating machine learning with asset management can help councils manage social housing allocation
1. Using historical data to help with prioritisation and allocation of social housing
In the last decade, there has been a total net loss of 165,000 social homes, equating to over 14,000 homes lost per year. This makes the process of prioritising and allocating homes a significant challenge. However, the integration of AI with asset management solutions can provide a powerful solution. An AI system can be trained on historical allocation data that is stored in asset management software and key criteria defined by the local council, such as; income level, family size, urgency and special requirements. Based on this criteria, the AI system can then score and rank applicants in an objective and unbiased way, ensuring those most in need are given priority. Further, AI’s advanced data processing capabilities allow for simultaneous consideration of multiple complex factors that may be impractical for manual review.
Moreover, AI can improve the matching process by finding the most suitable properties for applicants based on their unique requirements. This may include factors such as: proximity to schools, ground-floor access or proximity to transportation. The machine learning system, through analysis of data stored in asset management software, can also consider the current and future availability of properties, providing a real-time, dynamic matching system that exceeds the capabilities of manual methods.
2. Data analysis and predictions based on data stored in asset management software
AI, especially machine learning has the ability to analyse vast amounts of data to predict trends, understand patterns and make forecasts based on the data stored. How does this apply to local councils facing housing allocation challenges? Local councils handle vast amounts of data, ranging from applicant information, housing inventory, historical allocation data and maintenance records, all of which can be stored in asset management software. Processing all of this data manually to make informed decisions is often an imprecise and daunting task. Machine learning excels at finding patterns in complex data sets and making accurate predictions based on those patterns.
For instance, AI can analyse demographic trends, economic data and historical housing demands to predict future needs for social housing. This may include anticipating whether more houses are going to be needed to suit elderly people or family homes. Furthermore, AI can forecast which properties are likely to become vacant soon based on past tenancy durations, allowing councils to plan and allocate resources efficiently.
3. Preventive maintenance of local council assets
Machine learning systems can be used with asset management solutions to analyse historical maintenance data, housing conditions, weather patterns and other relevant factors to predict when and where maintenance is likely to be required in the future. For instance, AI can learn from past instances of when certain types of repairs were needed such as specific plumbing repairs that occur frequently in certain models of homes based on the data stored in the asset management software. This predictive capability allows councils to schedule maintenance proactively, before a minor issue escalates into a major problem.
As we step into a new era of asset management, local councils could benefit from understanding and exploring the opportunities AI and machine learning present for social housing allocation. The journey will require careful navigation and sufficient testing to ensure its implementation is suitable. The question is: Are local councils ready to embrace it?
Charlie Green is senior research analyst at Comparesoft.
