dvsa-logo.png

2018 

Brief: 

Bring to life DVSA’s (UK Driver Vehicle Standards Agency) digital vision of the future driving theory test

UX/UI designer

Role: 

UX (user journey), UI, demo video

Activities:

Screen Shot 2018-10-10 at 22.54.06.png

Background

We were approached to create a proof of concept that could enable people to take the test from any place, and on any device at any time whilst ensuring the authenticity and integrity of the test.

The problem

Currently when learning to drive in the UK, the first step is to complete the driving theory test, this test must be taken in person at a physical test centre. However, DVSA have an opportunity in 2020 to renew their agreement with the current test providers.

We conducted a collaborative 5 week proof of concept design sprint with DVSA to explore how theory test takers could take the test from home whilst ensuring the test integrity and authenticity remained. Prevention of fraud during the driving theory test is crucial due to the high stakes nature of the test however it is equally important to provide a truly brilliant customer experience at the same time.

The solution

A web-based solution that uses machine vision, type recognition and sound detection to enable users to take the driving theory test from home in an environment that adheres to the governments strict guidelines.

Test instructions

As so many hardware devices in the computer needed to be active in order for a user to complete the test, we decided that instructions to guide the user through the preparation process would provide a better user experience

07.png

Facial recognition

This is a visual to advise the user how to position themselves during the test. The camera built into the device, constantly detects the user's face, and flags up any suspicious behaviour

06.png

Surroundings

To ensure that test takers are not cheating, facial recognition technology also scans for other people who may be present to detect potential collusion

04.png

Object recognition

The camera not only detects faces, but also objects such as other devices and notebooks, and even water bottles - to ensure no notes can be used during the test

05.png

Sound

Sound is recorded to identify the test taker at the start and end of the test to authenticate the same person has answered the questions. The microphone can also detect any unusual background sound and flags potential cheating

03.png

Fullscreen

The test must be taken in full screen mode to prevent the test taker browsing another website and accessing material that would aid them. The second the fullscreen is exited, the test is marked a fail

01.png

Admin dashboard

I also designed a basic backend interface for DVSA employees to evaluate flagged tests, and decide whether fraudulent activity had occurred. The solution flags instances of activity that are considered 'suspicious' allowing the evaluator to skip through to each instance rather than have to watch the whole thing which could take up to almost an hour. 

The evaluator can quickly see the video recording of each incident and also hear soundbites where unusual noise was detected. 

At the end of the evaluation, a summary can be added and the case closed or escalated for further action.

02.png

Outcome

We presented our proof of concept back to the senior board of the DVSA's digital arm, with great success. We won further investment and then handed off the proof of concept to a delivery team within Capgemini who are taking this forward into a fully functioning platform that will be released in 2020.

Read another case study

To play, press and hold the enter key. To stop, release the enter key.