Combining vehicle data to create safer roads

Case study


Bringing together data from disparate systems helps identity the highest-risk vehicles

Police need to be able to identify the most high-risk vehicles on the roads – those which are uninsured, or which do not have a valid MOT or vehicle tax – but this data is held by separate organisations and not easily cross-referenced. 

National policing initiative Operation Tutelage compares automatic number plate recognition (ANPR) data with information from the Motor Insurance Database (MID) to identify drivers who do not have insurance. Tutelage Plus checks whether vehicles have a valid MOT certificate and whether Vehicle Excise Duty (VED) has been paid, through checks against the Driver and Vehicle Standards Agency (DVSA) and Driver and Vehicle Licensing Agency (DVLA) databases.  

The Department for Transport (DfT) and the Home Office are running the RADAR program which aims to further increase road safety by improving policing intelligence, and the Accelerated Capability Environment (ACE) was asked to explore what the art of the possible could look like in terms of bringing more vehicle data together in real-time. Advantages of this would include enhanced law enforcement capability, as drug-dealers typically used uninsured cars, as well as increased revenue for HM Revenue & Customs as more vehicles become compliant.  

Real-time updates 

The police maintain a list of thousands of Vehicles of Interest (VoI) created by aggregating data from the DVLA, DVSA and MIB data sets. This aims to prioritise non-compliant vehicles but is currently a manual process that is not done regularly, and which needs constant corroboration. 

The ambition is to make this real-time, so that a vehicle captured on, for example, an ANPR system can be compared to the most up-to-date data. As there are more than 10,000 vehicles on the VoI list at any one time, the DfT and Home Office also wanted a proof of concept architecture that will take in these data feeds and use some analytics to prioritise this list automatically in real time against dynamic rules which could be changed as the law or individual situations dictate.  

ACE together with suppliers from the Vivace community: 

  • Worked with DVLA, DVSA and the MIB on data discovery to determine the granularity of information available and how it could be ingested 
  • Produced a scalable, extensible and future proofed facsimile of the Tutelage programme which can incorporate disparate data sources to address non-compliant vehicles  
  • Produced a roadmap for future innovations that will improve the efficiency of the RADAR programme. 

Two additional strands of work looked at how ANPR data could be used to optimise police resource to put officers in the most effective place at the roadside while a data survey was also conducted to look at alternative data sources that could be used to augment reidentification in the up to 25% of cases on the VoI list where no registered keeper is known.

From: Accelerated Capability Environment