Exploring data standards and the UPRN workshop
Application of data standards would have many benefits to the fire and rescue service: sharing a common language, improved data quality, easier and quicker sharing of data, and integration of third-party datasets around common data points. Achieving these benefits would create opportunities for new trends and insight to be determined from larger and broader datasets.
The DDP is seeking to test this principle with the data analyst community, to explore how it could impact the fire and rescue service.
The UPRN is a unique identifier assigned to every addressable location in Great Britain and is maintained by GeoPlace, the custodians of the opensource database. It is freely available for use by public sector organisations and as of 01 July 2020 has been mandated as the standard by which the public sector share property and street information.
The ability to link datasets at the household level has the potential to unlock new trends and better identify and target risk and vulnerability in local communities, contributing to improved risk management and evidence-based policy decisions. A joint workshop was held with GeoPlace and data analysts from a cross-section of fire and rescue services to explore this potential.
It began with a review of how often building-fire related incident data in England had been submitted to the government IRS between 2010/11 to 2018/19 with the UPRN attributed. The application of the location standard varied greatly across services, though disparities in application could either be due to when the service started applying the UPRN to their datasets within the defined time period, and/or inconsistent application of the UPRN on the IRS - as it is not yet a required field for submission.
Figure 3: Fires in dwellings, other residential and non-residential buildings with a UPRN by fire and rescue service, England, 2010/11 to 2018/19
The analysts were asked to test the assumption that if a food outlet has a low hygiene rating then it is more likely to have a fire. This theory assumes that poor food hygiene management acts as an indicator of poor attitude toward managing other aspects of the business, including fire safety.
Focusing on two London Boroughs with a high density of food outlets, the analysts were provided with multiple datasets that include the UPRN, including fire incident data, inspection data (for non-domestic buildings), food hygiene rating scheme data from the Food Standards Agency, Energy Performance Certificate data and GeoPlace’s UPRN database product AddressBase. Analysts were split into groups and given a few hours to work with the data using any techniques they wanted and were challenged to ‘tell us something about fire risk that we don’t already know’.
You can read a full summary of the workshop on the GeoPlace blog.
While the results of the challenge were inconclusive, largely due to the small sample size, time available and lack of other third-party datasets to get a more rounded view of premises, the experiment takeaways reinforced how critical data quality is to data analysis and data linking, particularly when using one criterion - in this case the UPRN.
It also highlighted the importance of giving fire analysts the opportunity to put their heads up and look at fire data in a new way, with unfamiliar sources to stimulate new thinking. Getting out of the office and working with like minds on common problems is great for community building, engagement and knowledge sharing.