Jumping straight to CFD has become the default move in fire safety engineering, but that habit can quietly weaken our work: more inputs, more assumptions, more ways to be wrong, and often no clearer link to the actual design question. We sit down with Craig Hofmeister and Brian Klein to unpack a practical, defensible way to choose the right fire model for the job using the SFPE guideline “Substantiating a Fire Model for a Given Application.”

The broad framework of this work is to define the phenomena of interest and questions at hand, then choose the candidate models and evaluate them through set of core qualities, then address the verification and validation of the models, consider uncertainties and user impact, and finally document the whole process.

We walk through the framework step by step, starting where good performance-based design always starts: the questions the model must answer. From sprinkler and detector activation to atrium smoke control, pressurization, visibility and tenability, we talk about translating objectives into key physics and required outputs. That sets up a grounded comparison across hand calculations and algebraic correlations, zone models like CFAST, node network tools like CONTAM and Ventus, and field models like FDS built in PyroSim.

From there, we get into the part many projects rush past: verification versus validation, how to use published V&V evidence (and when you are outside the validated scope), and how uncertainty and user effects should shape your confidence. We also address real-world constraints like AHJ expectations and contract requirements, plus practical tools like sensitivity studies, bounding analysis, and grid sensitivity checks to keep complexity from turning into false precision.

If you want a cleaner way to defend your modeling decisions to reviewers and stakeholders, this conversation gives you a repeatable process you can build into your own practice.

----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.

“Two people per parking space” is one of those default fire engineering inputs that we are very used to place into a model without really thinking much of it. But it is one of those defaults that show a huge richness once you dig deeper. Are all parking spaces taken? Are people in their cars? What are they doing? How long have they been there concurrently... We take that simple rule and pull on the thread until it turns into a full conversation about evidence, uncertainty, and what “credible maximum” should mean when you are designing for real-world risk.

Dr Mike Spearpoint from the OFR joins me to explain how occupant load values end up in codes, why they are so hard to interpret, and why “maximum possible” can push designs into unrealistic corners. Then we get practical: we build a static, risk-based method for car park occupant load using distributions for car park utilisation and people per vehicle, run it through Monte Carlo simulation, and talk about selecting percentiles like the 95th or 99th for design. If you work with evacuation analysis, performance-based fire engineering, or fire safety assessment, this is the kind of reasoning you can reuse anywhere.

In his consideration, Mike reaches something he calls the dynamic model: people are only briefly “in the car park” as they park, unload, walk to the destination, and leave. Because published data on “around-the-car” activity time is scarce, Mike measures it directly using public CCTV observations and turns it into a usable distribution.

Why did he do this? This is a part of a larger project on adequate fire resistance periods in car parks. We also connect utilisation to vehicle-to-vehicle fire spread and why those assumptions can ripple into design fires and structural fire resistance decisions for open-sided car parks.

If you are looking for the report itself with all the details, look here: https://www.gov.uk/government/publications/fire-safety-open-sided-car-parks

I'll make it easy for you, it starts at page 218 ;)

----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.

A fire strategy can be technically correct, but if the team building the building never truly understands it - goals and objectives may be missed. For the 250th Fire Science Show, we slow down and talk about the craft of communicating fire science to construction professionals so that the intent survives real projects, real deadlines, and real handovers.

This episode is an extended version of my talk I gave recently at the IAFSS Research Sub-Committee Workshop, which we have organised with Felix Wiesner, and I had a chance to talk along my friends - prof. Guillermo Rein, Birgitte Messerschmidt and dr Steve Kerber.

In this episode, we share why the biggest failures are rarely tiny compliance misses. The scary failures come from misread strategy, missing execution on site, and teams optimizing for the wrong target because we explained the “what” but not the “why.” From smoke zoning misunderstandings to the way product labels and ratings get interpreted, we unpack how simple miscommunication can create life-threatening conditions even when everyone is working hard.

Then we offer a practical framework built around three ideas: context, timeliness, and the way we speak. Context means understanding the building ecosystem: code and local planning, sustainability and energy efficiency, LEED or BREEAM certification pressures, business model realities, and aesthetics. Timeliness means matching our message to the building lifecycle, keeping high-level objectives clear early on, translating them into technical concepts during design, and only then driving into the technical detailing that makes compartmentation, egress, smoke control, and structural fire safety real. Finally, we get honest about what works: simple anchors like ASAT versus RSAT, consequence-focused language, and respectful collaboration, plus what breaks trust fast: jargon, paper-style writing, megawatt talk, and false certainty around “60 minutes” ratings.

Some other podcast recommendations after this one:

• https://www.firescienceshow.com/136-fire-fundamentals-pt-6-the-fire-automation-in-a-building/ what happens in a building during a fire?
• https://www.firescienceshow.com/246-fire-fundamentals-pt-20-fire-resistance-criteria-with-piotr-turkowski/ a wider view on the fire resistance
• https://www.firescienceshow.com/199-commercial-timber-guidebook-with-danny-hopkin-and-luis-gonzalez-avila/ commercial timber guidebook which is an example of excellent communication of fire safety concepts.

----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.