The Digital Planning team in the Ministry of Housing, Communities and Local Government (MHCLG) has claimed two breakthroughs for the use of AI in the sector.
It said it has created a chatbot to pull out important information from local plans to answer questions, and developed a common way to classify and organise local plans.
The team has described the steps in a blogpost on the results of a six-week project it has recently run with AI specialist Faculty AI.
It said that, in a series of tests of its accuracy, the chatbot correctly answered questions how many homes were to be built within a specific timeframe 59% of the time, and on the lifetime of a local plan 76%. Answering questions on climate changed it provided 43% correct responses.
The classification system – otherwise known as a standard taxonomy – has been developed by the analysis of over 120,000 pages from 100 local plans using a large language model. This has led to the development of a standardised table of contents for the plans.
Navigating complexities
“These results show that AI has promise in navigating the complexities of planning documents and data, particularly given the time and effort usually required for such tasks, but highlights that the results are dependent on the quality of the input data,” Digital Planning said.
It is now working with local authorities through the PropTech Innovation Fund to see how AI can help with processing and organising feedback and consultation representations.
