The opportunity for data in fire safety

Following the publication of the FPA's 2024 Policy Manifesto, David Williams, chair of the FPA, examines why we need to build and make better use of data in planning system reforms

Support planning system reforms with data-driven fire protection measures

Policy aim: Support planning system reforms through the use of data to target fire protection measures effectively, particularly in high-risk buildings and those distant from fire and rescue services

Rationale: The planning system reform is widely considered overdue and essential to economic growth. The strategic use of sprinklers and suppression systems based on data can enhance life safety and economic benefits. This approach ensures that fire protection measures are applied where most needed.

Manifesto

Our manifesto, launched last year, included a policy that might sound less interesting than some other initiatives - to ‘support planning system reforms with data-driven fire protection measures’. Hopefully, however, we can convince you not only of the merits of supporting this approach, but also that this is an area with significant possibilities for even greater benefits going forward.

The simple position we take when discussing this area tends to be to ask “If you could use existing data to make better planning decisions, why on earth wouldn’t you? If you know a certain combination was more likely to lead to a fire, why wouldn’t you choose a different approach?”

Some might argue that planning regulations incorporate a bit of that already, decisions and rules are based around experience over time, but here we are talking about getting much more granular.

Data

At the FPA, we have a mass of information on buildings and large fires. Our Member Insurers already gain great insight from our Informer Database, designed to support risk control and insurance underwriting, but that is of course after a property is built and occupied.

Data is captured on the number of UK commercial and domestic fire claims, with details of fire claims by trade, industry, property, and outside area. Regional variations in fire claim numbers as well as trends in causes of fire ignition and factors contributing to any increase in fire-related claims are also collated.

Surely there would be benefit in planning system reforms, and individual planning decisions, using some of this information and being based on as much real data as possible?

In recent years, an increasing number of intelligent methods have been used for construction management and both construction sites and engineering projects having a more comprehensive, intelligent construction. This combined with the growth of the Internet of Things (IoT) and connected devices generally means we have more and more data that we can put to good use, and already some is used to improve building safety, albeit as previously mentioned more after a property is in use.

Internet of Things

IoT has revolutionised fire protection through the deployment of smart sensors in buildings. These sensors continuously monitor temperature, smoke, humidity, and air quality, providing real-time data to fire safety systems. IoT-enabled fire alarm systems can detect early signs of fire and send instant alerts to building occupants and emergency responders. And of course, all this information is recorded and kept – data that we currently don’t make full use of.

A mass of data is only the start, however. It’s great that we are recording things, and anecdotally we can see things that have contributed to a failure ‘after the event’. We need to make that data work for us however, and in the past, there have been difficulties in three areas: storage and accessibility of the data, the processing power needed to run models, and the cost of the expertise to see the patterns and predict outcomes.

Solutions

Things have moved on however, opening up much better possibilities for the future. Storage and accessibility is no longer the issue it was due to extensive adoption of ‘the Cloud’, with its secure storage, backup, ease of access, greater use by governments and businesses, and an increasingly affordable price. The solution to the cost of expertise problem is linked to the solution to processing power, as this grows exponentially, either in individual super computers or through the networking of thousands of individual machines. Part of the growth here has been driven by the need to facilitate what some people refer to as the next industrial revolution, Artificial Intelligence (AI).

Previously, data would be collected but sit unused because of a lack of experts/actuaries to analyse and investigate patterns. With AI and machine learning suddenly that problem disappears. Whilst I would always advocate the retention of human involvement, that need is now for the really interesting bits, selecting and fine tuning the recommendations from the algorithms, rather than having to do all that initial analysis manually.

So we have the data, the ability to make sense of it, and can use it to predict outcomes, now we need to start using that in the planning process.

New technologies, new risks

Despite advancements in fire protection technology, several challenges hinder effective fire safety in buildings and structures. Addressing these challenges requires planning system reforms that incorporate data-driven approaches.   

Many existing fire safety regulations were developed decades ago and have not kept pace with modern construction practices and materials. High-rise buildings, smart cities, and sustainable architecture introduce new fire risks that outdated codes fail to address.

Planning system reforms should integrate real-time data and predictive analytics into regulatory frameworks, ensuring that fire safety codes evolve based on new risks and technological advancements.

Predictive analytics and AI can play a crucial role in fire prevention by analysing historical fire data, environmental conditions, and building materials, to identify potential fire hazards. AI-driven models can assess risk factors such as occupancy types and levels, electrical loads, and material flammability, to predict fire outbreaks before they occur, and when used within the planning stage can suggest an alternative approach.

For example, machine learning algorithms can analyse patterns in fire incidents and predict high-risk areas. This data can be used by urban planners and fire departments to implement targeted fire prevention measures, such as improving building codes, enforcing safety regulations, and allocating fire response resources more efficiently.

We have used Building Information Modelling (BIM) for some time now, a digital representation of a building’s physical and functional characteristics. It allows architects, engineers, and fire safety experts to simulate fire scenarios and optimise building designs for fire protection.

By integrating BIM into the planning system, and adding the information produced from our new ability to understand and use existing data, authorities can ensure that new developments meet the highest fire safety standards before construction begins.

So, there you have it, the future gives us more meaningful data and more opportunities to use it, surely the case for supporting planning system reforms with data-driven fire protection measures, is one we can all agree we need to get behind?

About the author

David Williams, chair of the FPA

David is an experienced executive and non-executive director with over 30 years’ experience in insurance, having formerly held the position of Managing Director of Claims, Underwriting and Technical Services with AXA Insurance UK. He is a member of the FPA Board and between 2011-2022 was Chair of RISCAuthority, an annually funded scheme which conducts research on behalf of the insurance sector and is administered by the FPA

www.thefpa.co.uk