Back in the day, everyone knew someone who could get them a dodgy MOT on a not so road legal car, you know the type, that old banger with rust holes in the footwell the size of your hand. Take it to ‘Roy’ he’ll sort you out no problem, he won’t even need to see the car.
Shockingly that practice still goes on, it’s not as prolific thanks to the digitisation of MOT licenses, but it occurs, leaving potentially lethal cars on our roads.
But the DVSA are fighting back. Using the power of AI (Artificial Intelligence) they can now identify and target garages and testers that aren’t correctly testing vehicles. To do this, the DVSA worked up a profile of testers and stations with similar patterns, anyone that didn’t quite match this ends up in a high-risk category.
In doing so they’ve given individual testers a score and a record that moves with them between garages, it marks their performance and pass/fail rate. The system can also suggest changes to the behaviour of the tester over time as it learns their patterns, maybe they’re never flagging up worn tyres or blown bulbs?
By refreshing regularly, the algorithm keeps the database accurate and up to date, it also collects data on test volume, frequency, duration and disciplinary history.
All of this is then combined with the results of any prior enforcement action taken against the garage or tester, thus producing a risk rating.
It basically sifts out the bad garages and testers allowing the DVSA to send undercover officers to check out these nefarious traders.
But does it work?
So far, yes. In the last six months, the system has flagged far more cases than the DVSA would have found using their previous investigatory skills. In fact, the results have been so good that some garages were immediately banned from testing, with testers being given suspended prison sentences.
One such garage in Devon was found to have issued more than 300 fraudulent MOTs.
What’s next for AI and the DVSA?
As the system is so effective, the DVSA are working with their technology partner Kainos in developing the platform to identify where errors are being made in the MOT process.
This next step will profile vehicles and then predict when a failure will happen. If a tester contradicts this theorised result they can then be targeted and trained to spot the potential faults.
Aren’t MOT’s strict enough?
Last year saw the most extensive overhaul of MOT regulations in decades, from emissions, daytime running lights, brake fluid contamination to the removal of diesel particulate filters. Categories were also changed, instead of the ambiguous ‘Advisory’, there are now four categories for repairs.
Dangerous – Your car isn’t allowed on the road until the fault is fixed, driving your vehicle with a dangerous MOT rating will invalidate your insurance and could see you being stopped by the police.
Major – While not in the same league as dangerous, your car will fail its MOT and need the repair work to be carried out ASAP.
Minor – These faults need to be repaired as soon as possible, but don’t impact safety or the environment. Your car will pass.
Advisory – This category means you need to keep an eye on these issues, they could turn into potential faults in the future.
How do I know a garage is reputable?
Finding a decent, trustworthy garage can be a problem, especially if you’re new to the area and can’t rely on word of mouth. Don’t fear, thankfully here at Regit we maintain a database of MOT centres you can trust, just pop your postcode in to find your nearest.