New algorithms could soon help DVSA identify risky garages and testers

Agency researching ways it can better use its information for garage risk rating

New algorithms could soon help DVSA identify risky garages and testers
Image: Bigstock.

The DVSA plans to use mathematical algorithms to spot trends in garage or tester data that are worth investigating, it has revealed.

Describing the work as one of the main focuses for October, Neil Barlow, head of MOT policy at the DVSA, explained that data analysis may include volumes of tests in a given time, failure rates or items for the particular vehicles tested.

He added: “We would then combine this with other information to identify situations worthy of investigation, or just to determine a risk rating, which would be used to prioritise our work.

“We still have a lot of work to do on this, but will update you as we progress.”

The DVSA will also be changing the reasons for rejections and advisories this month to ensure motorists can easily understand them.

Latest system changes

In recent weeks, the DVSA added a link to the Society of Motor Manufacturers and Traders vehicle safety recalls page on the testing service.

It also added brake weights of more vehicles to the system.

Changes have been made to the test quality information part of the testing service, allowing users to choose either one or three months’ worth of data.

These include the number of tests, the average age of vehicles tested, the average test duration and failed test percentages.

Users can now see a breakdown of test failure component categories, including figures for the site averages.

Related: MOT “friendly” DPF removals condemned as garages set to face “serious consequences”

Do you support the DVSA’s recent work to improve the testing service? Share your comments and experiences in the comments below.

7 Comments

  1. Dvsa current system is a load of tosh. I fail more cars on incorrect headlight alignment – the amount that are incorrect is staggering. Because of this my failure rate is 68% (national average over last 3 months was 35%), and DVSA are on my case continually. I’ve had 4 re-inspections within the last year for myself alone.

    Reply
    • Yes you are correct. I see that in my fail sheets as well and on other issues, makes you wonder who records all.

      Reply
    • We test both cars and motorcycles and our pass rate is erratic and always has been in 25 years of testing. Why the heck do they not put more advisers on the ground to help garages, reopen training centres and shut down cheap mots? We could all start to make a profit and ensure better standards, the whole scheme is a mess and the public have no idea who to trust.

      Reply
    • Do what we do Paul. We adjust them FOC for customers. Takes 10 secs. And you don’t have to get them back in on the beam tester on a retest.

      Reply
      • How does that help as although you can do that during the test you should prs it and still record the failure

        Reply
  2. Not too sure about all this algorithm business, if a vehicle passes it passes if it fails it fails. Recording what fails correctly is more important than keeping close to any national average surly.

    Reply
  3. You will always get the garages who will not record a fail or Prs an item if it`s just a bulb blown or headlight alignment, I often get vehicles 10 + years old that we have not seen before and when you look into the history you can see they sometimes do not have as much as an advisory on them , this often creates a bad feeling between us and a new customer when the car fails on several items , especially when the customer says ” it went through okay at my last garage without any advisory’s”, makes us look bad but we do our testing to the book , if it does not meet the criteria set out in the testers manual fail it , if it`s close but safe pass and advise BUT RECORD EVERYTHING ! I have a high fail rate compared to the national average on some items but glad to say still a green ( at the moment ).

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>


The reCAPTCHA verification period has expired. Please reload the page.

Have your say!

0 0

Lost Password

Please enter your username or email address. You will receive a link to create a new password via email.