How the Department for Transport used AI to improve MOT testing
Learn how the Department for Transport (DfT) used clustering to ensure MOT standards remained high.
This guidance is part of a wider collection about using artificial intelligence (AI) in the public sector.
AI technique used
- clustering
Objective
The Driver and Vehicle Standards Agency (DVSA) wanted to use a more intelligent, data-driven approach to better target their resources and ensure MOT standards remained high.
Situation
Each year, 66,000 testers conduct 40 million MOT tests in 23,000 garages across Great Britain. A team of 300 examiners are responsible for ensuring garages are applying the MOT standards. Examiners assess garages every 1 to 3 years. This was resource intensive, and did not allow for targeted inspections.
Action
The DVSA applied a clustering model against garage test data from the last 3 months, as there was no labelled data available. The clustering model grouped MOT centres based on the behaviour they show when conducting MOT tests.
The DVSA created a risk score for each garage in Great Britain. The risk score combined the output of the clustering algorithm with historical data about how frequently garages had been disciplined for not applying correct MOT standards. This allowed the DVSA to rank garages and their testers, and helped the DVSA identify regional trends.
The DVSA updated their garage test data every 3 months. This allowed the DVSA to ensure the model used fresh data whilst also providing stability to garage ratings, providing examiners 3 month windows with which to visit garages.
Impact
The DVSA can now target their resources at the garages and testers with the highest risk score. By identifying areas of concern in advance, the examiners’ preparation time for enforcement visits has fallen by 50%.
There has also been an increase in disciplinary action against garages, meaning standards are now being better enforced. As more garages are delivering better MOT standards, there are more cars on the road that comply with roadworthiness and environmental requirements.
Related guides
- Understanding artificial intelligence
- Assessing if artificial intelligence is the right solution
- Planning and preparing for artificial intelligence Implementation
- Managing your AI project
- Understanding artificial intelligence ethics and safety
- Examples of real-world artificial intelligence use
- National Cyber Security Centre guidance for assessing intelligent tools for cyber security
- The Data Ethics Framework
- The Technology Code of Practice