A fire in an informal settlement is not just another small building fire. It can be the first domino in a fast-moving neighborhood event, and the little details like the wall material, roof material, door location, even a light breeze, can decide what happens next. I’m joined by Dr. Sam Stevens from Kindling to unpack a massive FSRI funded experimental program carried out in South Africa that burned twenty different informal and humanitarian shelter types to measure real heat flux, flame extension, and fire spread potential.

We start repeating the uncomfortable gap: humanitarian guidance often relies on rules of thumb like a universal separation distance and vague advice to “use fire-resistant materials,” with little experimental evidence behind it. Sam explains how Kindling built a global database of shelter designs and materials, then narrowed it down to common typologies spanning sheet metal, mud, timber, bamboo mats, split bamboo, thatch, and tarpaulin tents. We dig into the field-scale test setup, why wood cribs were chosen, and how external instrumentation like thin skin calorimeters helps quantify the heat transfer that drives structure-to-structure ignition.

Then we get into what the burns actually show. Non-combustible shelters often behave like classic compartment fires where openings dominate, but even modest wind can push flames meters outward. Add combustible walls or roofs and the hazard shifts dramatically: some materials produce short, intense exposure windows while others sustain burning long enough to threaten neighbors at greater distances. The fully thatched shelter stands out for extreme radiant heat and duration, and our discussion on tarpaulin tents reveals why common fire test methods can produce reassuring ratings that still miss real-world behavior.

If you care about informal settlement fire safety, humanitarian shelter design, fire engineering guidance, or modeling large outdoor fires in dense communities, this conversation gives you a data-driven foundation and a clear sense of what questions still need answering.

Learn more about the project and watch their educational videos here: https://kindlingsafety.org/projects/large-scale-fire-experiments-for-humanitarian-shelters/

----
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.

Being a part of broader civil engineering and built environment sciences, we have the unique opportunity to learn from other "sister" disciplines, rather than coming up with everything on our own. Especially, when those disciplines have 100+ years of experience in investigating stuff that has recently emerged as one of the leading challenges in our field.

The other discipline is Earthquake Engineering.

The interesting stuff - community resilience and managing safety across tens of thousands of buildings exposed to the same hazard.

Our field of application - Fire Safety of WUI communitites.

And most importantly, my diligent guests: Negar Elhami-Khorasani from University at Buffalo, and Justin Moresco from Applied Technology Council.

In this episode we discuss how earthquake engineering deals with hazards through four steps:

1. Hazard assessment,
2. System response (exposure)
3. Damage assessment (vulnerability)
4. Level of loss (consequences)

You will learn what are fragility functions and how they are applied. We will also discuss building archetypes and how we can use them to increase or decrease the level of detail in our analysis.

If you want more:

• Conceptualizing a probabilistic risk and loss assessment framework for wildfires
• Toward Probabilistic Risk Assessment of Wildland–Urban Interface Communities for Wildfires

Recommended follow-up podcast episodes:

• 156 - Trigger Boundaries with Harry Mitchell and Nick Kalogeropoulos
• 215 - Lessons from the 2018 Camp Fire with Eric D. Link
• 222 - Integrating WUI risk management and fire safety engineering with Pascale Vacca

----
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.

Is it too late to start with the AI in 2026? It wen't so far, does it still make sense to get interested in this technology?

Absolutely. Today we sit down with MZ Naser of Clemson University to map a clear, useful path for engineers who want results without the hype. We start with the basics - clean data, the right algorithm, and a realistic mindset - and climb toward explainability, causality, and even philosophy to show where AI informs decisions and where it can quietly mislead.

We dig into the limits of our experiments: when tests are expensive, we control only a few variables and then celebrate when explainable AI “finds” the same drivers. That’s not discovery; that’s confirmation. MZ explains how broader sampling, anomaly detection, and careful clustering can reveal patterns we miss, while acknowledging that physics is fixed but our datasets are narrow. We also talk scale: a model that predicts whole-building fire behavior from scratch is a fantasy without impossible data. The practical play is combining reasoning, physics, and simulation to guide where AI adds value - sometimes leading to a simpler equation that replaces the model altogether.

Then we get tactical. What is agentic AI, and how can it save engineers real time? Think delegated workflows: data gathering, parametric setup, code lookups, Excel design sheets, quality checks, and concise summaries. Train agents with explicit steps and tight guardrails, keep them away from money and safety-critical controls, and make human review mandatory. We also confront traceability and model retirement - why freezing working versions, documenting assumptions, and cross-verifying with independent methods matter for audits years down the line.

Throughout, we balance open local models versus cloud LLMs, the trade-offs between control and convenience, and the hard truth that black boxes don’t absolve us of understanding. The big takeaway: AI is a lever, not a miracle. Use it to widen your view, automate routine work, and challenge your priors - while keeping physics, data quality, and professional judgment at the center.

If this conversation helps you think clearer about where AI fits in your workflow, follow the show, share it with a colleague, and leave a quick review so more engineers can find it.

----
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.