March 25, 2026

244 - Decision making in large-scale evacuations with Erica Kuligowski

244 - Decision making in large-scale evacuations with Erica Kuligowski
The player is loading ...
244 - Decision making in large-scale evacuations with Erica Kuligowski
244 - Decision making in large-scale evacuations with Erica Kuligowski
Apple Podcasts podcast player badge
Spotify podcast player badge
Apple Podcasts podcast player iconSpotify podcast player icon

When one takes a decision to evacuate and starts moving, this is not the end of their decision-making process. 

Which route to take? Who to contact? How to arrange a place of shelter? Where to go first? Have I forgotten anything? 

I previously discussed the decision-making with Erica Kuligowski from RMIT, and today we're meeting again to follow up on decision-making for large-scale evacuations. We focus on choices and uncertainties that make many of the evacuees take additional trips, and those trips become background traffic that interferes with your escape. In this episode, we dive deep into the decision making in this stage, the sources of data, and hypothesise how this knowledge could be used in practice.

And of course, Erica being one of the leaders of the Human Behaviour in Fire community gives us a high level overview how this part of science looks like, and what is currently being researched.

The HBiF conference we mentioned in the episode can be found here: https://humanbehaviourinfires.se/

----
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.

WEBVTT

00:00:00.299 --> 00:00:00.869
Hello everybody.

00:00:00.869 --> 00:00:02.160
Welcome to the Fire Science Show.

00:00:02.160 --> 00:00:11.369
This week we're venturing in the world of, uh, human behavior in fire, and I have an absolutely excellent speaker to discuss this, uh, important matter with.

00:00:11.369 --> 00:00:14.490
And that is Erica Kuligowski from RMIT.

00:00:14.490 --> 00:00:23.600
You may remember an episode with Ika that was in the early days of the podcast where she was moving into Australia, and we were discussing, uh, the decision making.

00:00:23.600 --> 00:00:25.181
Fires.

00:00:25.181 --> 00:00:34.780
Interestingly, some years later she's still working on decision making fires, and today we can discuss it even more.

00:00:34.780 --> 00:00:41.320
And actually we, we discuss a very specific aspect of decision making, you know, um.

00:00:41.320 --> 00:00:48.106
From the larger image in fires we have those things that, very common, that are quite simple.

00:00:48.106 --> 00:00:55.485
You know, when you wrap them in statistics, like in case of evacuation, the pre evacuation time distributions, for example.

00:00:55.485 --> 00:01:03.375
But once you want to have a detailed model of how something works, it becomes extremely complicated.

00:01:03.375 --> 00:01:06.436
And without those complexities, you cannot really model.

00:01:06.436 --> 00:01:08.021
It accurately.

00:01:08.021 --> 00:01:21.986
The same case is for decision making and vacation processes, and in this case, the specific niche that Eric is interested in and that we mostly discuss in today's episode is what happens.

00:01:21.986 --> 00:01:25.206
Between the decision to evacuate.

00:01:25.206 --> 00:01:35.587
So a person takes a decision, they want to evacuate, and, them actually reaching a place of safety because it's not that they teleport or venture always in a straight line.

00:01:35.587 --> 00:01:38.811
A lot of things happen and those things influence how others.

00:01:38.811 --> 00:01:48.231
Can evacuate those things influence the capacity of road networks and, uh, those things will influence how the evacuation process will happen at large.

00:01:48.231 --> 00:01:52.462
Something we don't really consider that much, but perhaps we should.

00:01:52.462 --> 00:02:01.581
This is of course, carried out in relationship to large scale evacuations, but a lot of concepts that are discussed are highly relevant for buildings.

00:02:01.581 --> 00:02:07.867
So if you are a building person, I would still recommend you to listen because Erica is, uh, just a brilliant speaker.

00:02:07.867 --> 00:02:13.883
And also she has this unique overview of the entire field of human behavior in fire.

00:02:13.883 --> 00:02:17.873
So at the same time, it's kind of a review what the field is doing.

00:02:17.873 --> 00:02:19.644
I hope you will enjoy it.

00:02:19.644 --> 00:02:21.054
I have enjoyed a lot.

00:02:21.054 --> 00:02:23.514
Let's spin the intro and jump into the episode.

00:02:47.338 --> 00:02:51.568
The Fire Science Show podcast is brought to you in partnership with OFR Consultants.

00:02:51.568 --> 00:03:01.340
OFR is the UK's leading independent multi-award winning fire engineering consultancy with a reputation for delivering innovative safety driven solutions.

00:03:01.340 --> 00:03:10.568
we've been on this journey together for three years so far, and here it begins the fourth year of collaboration between the Fire Science Show and the OFR.

00:03:10.568 --> 00:03:27.995
So far, we've brought you more than 150 episodes, Which translate into nearly 150 hours of educational content available, free, accessible, all over the planet without any paywalls advertisement or hidden agendas.

00:03:27.995 --> 00:03:34.985
This makes me very proud and I am super thankful to OFR for this long lasting partnership.

00:03:34.985 --> 00:03:50.865
I'm extremely happy that we've just started the year four, and I hope there will be many years after that to come So big thanks, OFR for your support to the Fire Science show and the support to the fire safety community at large that we can deliver together.

00:03:50.865 --> 00:03:58.423
And for you, the listener, if you would like to learn more or perhaps even become a part of OR, they always have opportunities awaiting.

00:03:58.423 --> 00:04:02.862
Check their website@orconsultants.com And now let's head back to the episode.

00:04:03.832 --> 00:04:04.462
Hello everybody.

00:04:04.462 --> 00:04:07.703
I am joined today by Erica Kuligowski from RMIT.

00:04:07.703 --> 00:04:08.362
Uh, hey Erica.

00:04:08.362 --> 00:04:09.712
Welcome back to the podcast.

00:04:10.694 --> 00:04:12.074
Thank you so much, EK.

00:04:12.074 --> 00:04:13.604
It's nice to be back.

00:04:14.002 --> 00:04:16.012
to invest 200 something episodes.

00:04:16.012 --> 00:04:17.843
It's absolutely crazy.

00:04:17.843 --> 00:04:20.483
Last time we talked you were moving to Melbourne.

00:04:20.483 --> 00:04:23.894
Yes, yes, it, that's been a while.

00:04:23.894 --> 00:04:25.658
So There's there, there's

00:04:25.694 --> 00:04:26.144
happened.

00:04:26.273 --> 00:04:30.622
there, there's a lot to catch up, but I'm even more excited to, to do this here.

00:04:30.622 --> 00:04:39.588
Um, observing what you're doing nowadays in Australia, you've ventured, uh, pretty far away from our good old building, uh, fires.

00:04:39.588 --> 00:04:45.000
Like what's the field of, uh, human behavior in fire that you're most interested in today?

00:04:45.245 --> 00:04:50.165
Yeah, so looking back on my career, I have, yeah, you're right.

00:04:50.165 --> 00:04:59.418
I've, I've sort of, moved from the building fire space more into the wildfire and bushfire space, and I did that a little bit before I had moved to Australia.

00:04:59.418 --> 00:05:24.447
But since I've been here, I've been very much working in, uh, the large scale hazards and disasters space, understanding how people respond different types of hazards and uh, trying to understand how we can improve emergency communication and also evacuation strategies for in hazard prone areas.

00:05:24.447 --> 00:05:27.267
So I've been looking at wildfires or bush fires.

00:05:27.267 --> 00:05:35.331
I have, uh, a few students who are PhDs working in the tsunami space and then I also have been working in floods as well.

00:05:35.331 --> 00:05:37.372
So lots of work and very

00:05:37.699 --> 00:05:38.744
And very large scale.

00:05:38.744 --> 00:05:41.019
okay, uh, you've ventured into Australia.

00:05:41.019 --> 00:05:43.240
What's the difference between the wildfire and bushfire?

00:05:43.240 --> 00:05:45.250
Is it like inherently different thing?

00:05:45.250 --> 00:05:48.639
I don't think I have the bushfire episodes specifically in the podcast yet.

00:05:48.793 --> 00:05:50.564
No, so it is very similar.

00:05:50.564 --> 00:05:56.430
It's a, it's just the term that we use here we don't use wildfire as the term, it's mainly bush fire.

00:05:56.430 --> 00:06:06.505
but we here in Australia call sort of the forests or heavily dense areas, the bush, so heavily vegetation dense areas.

00:06:06.505 --> 00:06:08.875
So that's where the bush fire term comes from.

00:06:08.875 --> 00:06:13.665
but if in fact it is a grass fire instead, then we'll use that term.

00:06:13.665 --> 00:06:21.961
So there is certainly, Differences in the fires that we have here, similar in to the US and Canada and elsewhere.

00:06:21.961 --> 00:06:25.740
But yes, we pretty much have wildfire synonymous with bushfire.

00:06:26.454 --> 00:06:31.269
I hope you are not developing something like Buoy Bush Urban Interface.

00:06:32.341 --> 00:06:35.651
No, no, no, no, no, no, no, no, no.

00:06:35.651 --> 00:06:37.750
We don't really use the term wooey here.

00:06:37.750 --> 00:06:46.180
Mainly bushfire prone areas or areas on the urban fringe or peri-urban areas that are, um, at risk of bushfire.

00:06:46.204 --> 00:06:47.103
That, that's also nice.

00:06:47.103 --> 00:06:48.004
I, I like that.

00:06:48.004 --> 00:06:48.569
I like, I,

00:06:48.911 --> 00:06:49.240
Yeah.

00:06:49.473 --> 00:06:51.303
I'm not sure if I'm happy with the woo term.

00:06:51.303 --> 00:06:54.209
I just find it really difficult to say, but.

00:06:54.761 --> 00:06:56.050
Doesn't roll off the tongue.

00:06:56.050 --> 00:06:56.411
No.

00:06:57.213 --> 00:06:57.454
Yeah.

00:06:57.454 --> 00:06:58.293
But, but yeah.

00:06:58.293 --> 00:07:04.701
Nothing better comes to my mind Um, and anyway, let's talk about those, uh, large scale, um, fire evacuation.

00:07:04.701 --> 00:07:09.471
So, uh, last time you were here and I highly recommend everyone to listen to that episode.

00:07:09.471 --> 00:07:12.442
It's all is still very, very highly relevant.

00:07:12.442 --> 00:07:13.401
Link is in the show notes.

00:07:13.401 --> 00:07:29.809
last time we were here, we were discussing your PADM model, uh, model that allows us to, to kind of model the behavior of, of people when they evacuate or the actions that people are taking, from the moment they, they see the fire to the moment they, they evacuate.

00:07:29.809 --> 00:07:36.048
Um, is this something that still is in the core of the development when you transition into large scale fires?

00:07:36.048 --> 00:07:37.908
Is does it work in similar way?

00:07:38.115 --> 00:07:39.014
Absolutely.

00:07:39.014 --> 00:07:46.689
So yeah, I used, um, Lindell and Perry's, PADM, action decision model in a lot of my building fires work.

00:07:46.689 --> 00:07:51.759
But it actually was developed, uh, very much out of large scale hazards.

00:07:51.759 --> 00:07:53.348
And so that's what it's based on.

00:07:53.348 --> 00:07:58.209
Um, and so it was sort of bringing that disaster lens to building fires.

00:07:58.209 --> 00:08:08.499
And so when I moved over to Wildfire research, it is absolutely at the core of what I do when I'm studying evacuation decisions.

00:08:08.499 --> 00:08:17.245
So, 100% it is a, a model that I use to develop the questions that I ask when I talk to bushfire survivors.

00:08:17.245 --> 00:08:25.975
And then it's certainly a lens that I use when I try to understand what people, uh, were doing, how did they decide to evacuate or not.

00:08:25.975 --> 00:08:29.690
But I've sort of moved a little bit from that, research, um.

00:08:29.690 --> 00:08:34.395
Not just looking at evacuation decisions, but looking at a few other things.

00:08:34.815 --> 00:08:35.669
Which are,

00:08:36.107 --> 00:08:46.638
So a lot of my work in the last, I would say three years, has been looking at what people do after they decide whether or not they evacuate.

00:08:46.638 --> 00:08:53.327
And so there's quite been quite a bit of work, and it was actually a keynote that I gave at the IFSS during the pandemic.

00:08:53.327 --> 00:09:04.104
that there has been a lot of work already in Australia and Canada and the US and abroad, um, on what influences people to evacuate or not.

00:09:04.104 --> 00:09:11.364
What is of a focus in our research is what people do after

00:09:11.427 --> 00:09:11.846
Mm-hmm.

00:09:11.964 --> 00:09:12.504
decision.

00:09:12.504 --> 00:09:15.894
So that's where I've been focusing a lot of my research.

00:09:15.894 --> 00:09:21.864
So it has been, and I'm presenting on this at the next IFSS, which I'm very excited.

00:09:21.864 --> 00:09:26.394
What are the factors that influence people's decision?

00:09:26.394 --> 00:09:28.453
Of where they go for safety.

00:09:28.453 --> 00:09:38.443
So a lot of times we know, um, from some of the research that's out there, actually a lot of it is in hurricanes in the us uh, that people go to hotels or motels.

00:09:38.443 --> 00:09:47.653
They can go to an evacuation shelter, they can go to a friends and family home, they could go to a caravan park, they could go somewhere else.

00:09:47.653 --> 00:09:52.687
They could maybe stop at a parking lot one night if they're really in a, an urgent situation.

00:09:52.687 --> 00:09:59.976
But what we really want to know from this research is where do people go for safety and why?

00:09:59.976 --> 00:10:05.753
The other questions that I'm asking as part of this research is, how do people get there?

00:10:05.753 --> 00:10:08.182
So do they ride share?

00:10:08.182 --> 00:10:11.153
And that's a lot of Steven Wong's work actually in Canada.

00:10:11.153 --> 00:10:13.105
do they take their personal vehicle?

00:10:13.105 --> 00:10:14.846
Do they take multiple vehicles?

00:10:14.846 --> 00:10:15.655
Are they.

00:10:15.655 --> 00:10:22.030
a trailer along with them, and maybe they've got their horses, um, that they're, that they're traveling with.

00:10:22.030 --> 00:10:25.660
We're trying to understand where do people go for safety?

00:10:25.660 --> 00:10:27.130
How do they get there?

00:10:27.130 --> 00:10:33.941
In what mode of transport and, and what route are they taking on the roadways, um, to get there.

00:10:33.941 --> 00:10:39.005
So it's a lot of work and it's been really, really interesting in what we've been finding.

00:10:39.443 --> 00:10:42.798
What's the ultimate relevancy of that, or practical relevancy of that?

00:10:42.798 --> 00:10:54.797
I mean, in building evacuation, if I think about the, the, the stuff that happens after a person decided, I mean, I would broadly put it in my pre evacuation time, distribution term.

00:10:54.797 --> 00:10:56.027
So for me.

00:10:56.027 --> 00:11:00.312
It kind of ends up into it a delay factor.

00:11:00.312 --> 00:11:02.711
I mean, I wish I would model that better.

00:11:02.711 --> 00:11:13.138
I wish I could include for way finding and you know, the fact that we're doing today, the visibility work, perhaps, you know, those things couple together beautifully.

00:11:13.138 --> 00:11:32.008
It's just not yet feasible to do that, uh, kind of work where you could combine, if you can see the sign and, and what decisions do you take and what we are finding to take, uh, regardless, like I, I, I would be looking for a specific time delay to put in my model because I'm ultimately interested in a AOT

00:11:32.277 --> 00:11:32.496
Yes.

00:11:32.799 --> 00:11:34.210
in wildfire.

00:11:34.210 --> 00:11:35.559
How does it look like

00:11:35.798 --> 00:11:36.038
yeah.

00:11:36.038 --> 00:11:37.988
I mean, that's a, a really great question.

00:11:37.988 --> 00:11:38.798
Why do we care?

00:11:38.798 --> 00:11:39.758
Why, why do we care

00:11:39.855 --> 00:11:40.995
this is important?

00:11:40.995 --> 00:11:41.559
Why?

00:11:44.258 --> 00:11:45.217
Why am I doing this?

00:11:45.217 --> 00:11:46.988
Yeah, that's right.

00:11:46.988 --> 00:11:49.149
And so You're absolutely right.

00:11:49.149 --> 00:11:53.860
There could be back of the envelope calculations that people make when we're thinking about wildfire.

00:11:53.860 --> 00:12:02.994
a lot of the assumptions that we make when we're thinking about how long people will take to evacuate a community are pretty optimistic assumptions.

00:12:02.994 --> 00:12:23.453
So we assume, for example, in the models that we use, or in the back of the envelope calculations for just traffic calculations that we make, that people go from point A where they start directly to point B and they take, a lot of times we assume that they take the fastest route or the quickest option.

00:12:23.453 --> 00:12:26.135
Yeah, that's right.

00:12:26.135 --> 00:12:28.686
Whatever Google Map says is the quickest way to get there.

00:12:28.686 --> 00:12:31.056
That's, that's what we assume people will take.

00:12:31.056 --> 00:12:37.206
maybe we assume that they will split themselves equally across the road networks.

00:12:37.206 --> 00:12:41.706
Um, and that also helps with quicker evacuation of a community.

00:12:41.706 --> 00:12:45.666
And we also assume that they don't come back in with that.

00:12:45.666 --> 00:12:52.769
They're actually just going directly to safety and they're not taking additional trips and a lot of these assumptions.

00:12:52.769 --> 00:13:02.130
I I think the last one that I'll mention is that in a lot of the models that we use, we assume a particular percentage of background traffic.

00:13:02.130 --> 00:13:06.061
And so those are people who aren't going directly to safety.

00:13:06.061 --> 00:13:07.681
They could be driving through.

00:13:07.681 --> 00:13:14.431
The problem is we don't have a lot of data on how much back background traffic should we assume.

00:13:14.431 --> 00:13:19.134
So it's really up to the user and then also with the.

00:13:19.134 --> 00:13:26.400
optimistic assumptions, we could be wildly underestimating how long a community takes to evacuate.

00:13:26.400 --> 00:13:30.870
Uh, people aren't always going the quickest route.

00:13:30.870 --> 00:13:38.461
They are likely to go a familiar route or a route that may include back roads because they're trying to avoid roadblocks.

00:13:38.461 --> 00:13:41.791
Um, we also know that they don't go directly to safety.

00:13:41.791 --> 00:13:46.591
We know that they're doing a lot of other things within the community and traveling around on the roadways.

00:13:46.591 --> 00:13:58.081
And when we're not taking that into account, or even when we don't understand what that looks like, a we can't assign a, a delay to it because we don't quite know yet what it is.

00:13:58.081 --> 00:14:03.567
And if we just assign a delay to it, we may not really truly.

00:14:03.567 --> 00:14:10.782
Be modeling the intricacies of what's happening and potentially underestimating the total evacuation time.

00:14:10.782 --> 00:14:22.631
So I am trying to help us get a little bit closer to what's actually happening on the roadways, so we have a better idea of how long a community will actually take to evacuate.

00:14:23.027 --> 00:14:35.241
So in that case, it's both the capacity of the roots and also how much stress that evacuation will put on the, uh, firefighters and, and the policemen and everyone around.

00:14:35.241 --> 00:14:35.542
Right.

00:14:35.841 --> 00:14:36.860
That's exactly right.

00:14:36.860 --> 00:14:39.500
So it's definitely a capacity on the roadway issue.

00:14:39.500 --> 00:14:45.801
So if people are traveling around, they're coming back to their homes, they're visiting neighbors, they're evacuating together,

00:14:46.072 --> 00:14:46.192
I.

00:14:46.370 --> 00:14:54.831
means that they're still on the roadways in fire affected areas, and there isn't as much capacity available for people moving directly to safety.

00:14:54.831 --> 00:14:58.370
We also need to take into account, as you said, how.

00:14:58.370 --> 00:15:03.471
The emergency services are using the roadways where they may be setting up roadblocks.

00:15:03.471 --> 00:15:05.390
What does that mean for evacuation?

00:15:05.390 --> 00:15:25.071
And the other thing that we're sort of on the tip of the iceberg for understanding, and I wanted to highlight, um, Constanza Flores, who is a PhD student at University of Canterbury in, in addition to RMIT, she's looking at, uh, potentially in tsunami, but it's very relevant for fires.

00:15:25.071 --> 00:15:30.650
People who are in heavy congestion, potentially leaving their cars and walking.

00:15:30.673 --> 00:15:30.832
Okay.

00:15:31.160 --> 00:15:41.484
And so that's another thing is because we potentially may have vehicles on the roadways for various reasons, and car abandonment may be one of them, accidents may be another.

00:15:41.484 --> 00:15:51.413
And that could be additional capacity taken up that we're not accounting for in any of our models part, particularly when we're trying to model worst case scenarios.

00:15:51.782 --> 00:16:06.455
This is very interesting, especially like, you know, the aspect of, of people coming back if I, again, normalize it to my building scenario and you know, I have my capacity of staircase basically.

00:16:06.455 --> 00:16:12.475
complete, you know, there it is just filled with people and someone decides, oh no, I, I want to go back five floors up.

00:16:12.475 --> 00:16:17.365
This creates a, a, like, not just a bottleneck, it's even more than a bottleneck.

00:16:17.365 --> 00:16:21.865
It's like a shockwave, uh, through, through the, you know, ongoing process.

00:16:21.865 --> 00:16:27.414
I, I, I struggle to imagine how a person could return back against a stream of evacuating people.

00:16:27.414 --> 00:16:41.804
I, assume if you have an endless stream of vehicles and people escaping, someone trying to, you know, go the other way, could, could also create such effects and those could propagate in the network.

00:16:41.993 --> 00:16:42.113
Y.

00:16:42.113 --> 00:16:43.207
Absolutely.

00:16:43.207 --> 00:16:47.227
And what we need to think about is not necessarily always people going back.

00:16:47.227 --> 00:16:57.802
These are people sort of from their original point, they're moving around and so they're, they're moving on the road network but not necessarily directly to safety.

00:16:57.802 --> 00:17:11.051
And so we need to sort of think about, they're on the roadways, potentially visiting a neighbor and then maybe they visit another neighbor and maybe they go back to, you know, help someone out and maybe they go and check on someone else's pets.

00:17:11.051 --> 00:17:19.271
And so it's sort of this movement around the road network within fire affected areas that we don't have a good sense of.

00:17:19.271 --> 00:17:20.592
So it might be that.

00:17:20.592 --> 00:17:24.343
Some areas have congestion, maybe some areas don't.

00:17:24.343 --> 00:17:37.452
But when we assume that people go directly from point A to B, we're not taking into account this other sort of, these flows that are going on within fire affected areas that are taking up capacity on the roadways.

00:17:37.452 --> 00:17:41.930
And the other thing that we see our people, like you said, are coming back.

00:17:41.930 --> 00:17:53.390
So they might be going to a location, dropping off some, a vehicle, coming back, getting another vehicle, and then coming back and maybe bringing, getting their pets and evacuating for good.

00:17:53.390 --> 00:17:57.377
And so that also could be, you know, some trouble.

00:17:57.377 --> 00:18:03.948
But we're, we're not necessarily thinking of complete congestion, not like we would have in a stairway in a building.

00:18:03.948 --> 00:18:14.147
This isn't always the case, but again, it's this sort of flow of people moving around rather than from fire affected area out.

00:18:14.147 --> 00:18:18.258
It's like, you know, moving, moving around and.

00:18:18.258 --> 00:18:22.746
creating other dynamics that we're actually not modeling in, in real

00:18:22.823 --> 00:18:37.146
Okay, but, well, this is really, really complex to grasp because let's assume you have evacuated community and you know that George went to Josh and had to pick, uh, his cat because he's outside and you mapped his interactions and you knew who went where.

00:18:37.146 --> 00:18:44.073
You can probably model that, but if you are on a blank page, you have a community of 1000 people.

00:18:44.589 --> 00:18:44.799
Hmm.

00:18:44.877 --> 00:18:47.553
How, how, how are you supposed to include for that?

00:18:48.055 --> 00:18:48.355
Yeah.

00:18:48.355 --> 00:18:51.144
So we don't have enough data to do that yet.

00:18:51.222 --> 00:18:51.603
Okay.

00:18:51.724 --> 00:18:56.515
we're trying to understand are the factors that influence that kind of movement.

00:18:56.515 --> 00:19:01.111
And I do have, one paper that I've developed from our interviews.

00:19:01.111 --> 00:19:05.736
there's been some work also from some Fire, A fire in Israel as well.

00:19:05.736 --> 00:19:16.227
Um, where they have this is At least five years ago, I believe they've done this work, um, to try to understand these intermediate trips is what we're

00:19:16.278 --> 00:19:16.699
Mm-hmm.

00:19:17.126 --> 00:19:47.156
So I think once we have a better sense of what are the factors that influence these types of trips and whether or not they're trips within a fire affected area or trips outside that come back in, then I think we'll be able to at least identify some percentage of background traffic that is leads based on a little bit more reality than what we have now, which is zero data, um, to, to try to get at some of these capacity issues.

00:19:47.156 --> 00:19:55.864
I don't think we're ever gonna get at mapping exactly what everyone's doing at any one time, we need to be able to get a little bit closer.

00:19:55.864 --> 00:19:59.013
And I think we, once we have a sense of, okay.

00:19:59.013 --> 00:20:04.365
This is a stronger knit community, so they're more likely to do a lot of internal movement.

00:20:04.365 --> 00:20:08.938
And this is a, community that has stronger connections.

00:20:08.938 --> 00:20:14.067
I don't know, 20 Ks away, but, so they're more likely to do a little bit of back and forth.

00:20:14.067 --> 00:20:25.121
Then we can start to maybe get a little bit more accurate with at least modeling our background traffic and trying to get at implicitly modeling some of these intermediate trips.

00:20:25.679 --> 00:20:41.523
I, I once, you know, I once thought that Orwell is 1984 is science fiction, but now I see it's more like a manual and I, my mind is in, in quite, uh, you know, sad, uh, prediction for a very dystopian future.

00:20:41.523 --> 00:20:44.733
So we might actually be able to do what you just said, you know,

00:20:45.185 --> 00:20:45.405
Yes,

00:20:45.634 --> 00:20:47.223
but, uh, perhaps,

00:20:47.306 --> 00:20:48.306
I, I know what you're saying.

00:20:48.306 --> 00:20:48.586
Yes.

00:20:48.693 --> 00:21:04.096
but, but perhaps at large, you know, just knowing the probabilities of how many people would do those side trips would give you enough to, it kind of, it kind of looks like a turbulence problem in a flow, you know, like you have, uh, or a gust problem in the wind.

00:21:04.096 --> 00:21:10.936
You have your average, uh, wind velocity and you have some gusts, which, which, which will change a little bit every now and then.

00:21:10.936 --> 00:21:12.767
So, so I, I like it a lot.

00:21:13.239 --> 00:21:27.989
this is why I love speaking with very curious engineers like yourself, because sometimes I'm very much in the social science space and so I wanna learn and dig in and then you'll sit and ask me, yeah, okay, but how do we, how are we gonna model that?

00:21:27.989 --> 00:21:31.888
So I love, that's what I love about interdisciplinary research and

00:21:31.977 --> 00:21:32.057
Hmm,

00:21:32.669 --> 00:21:38.398
teams is because you need both of those sort of minds thinking in that way.

00:21:38.398 --> 00:21:41.368
Like, okay, I'm going to get into the social sciences of this.

00:21:41.368 --> 00:21:49.259
I really wanna understand what people are doing and why, and then let's sit and think about how are we actually gonna calculate, quantify this, I mean to say.

00:21:49.259 --> 00:21:50.519
So I love that.

00:21:50.632 --> 00:21:56.271
I'm an engineer, but I, I built up enormous respect for social sciences.

00:21:56.271 --> 00:21:57.771
I eventually.

00:21:57.771 --> 00:22:01.616
Admitted that we are solving a human problem.

00:22:01.616 --> 00:22:07.616
So if we're solving a hu, if we are solving a human problem, you, you have to have human inside the problem.

00:22:07.616 --> 00:22:14.727
You cannot like solve it for a machine unless you are, uh, solving for evacuation of, uh, a robot army or whatever.

00:22:14.727 --> 00:22:17.366
Uh, but I don't, uh, care about that kind of evacuation.

00:22:17.366 --> 00:22:25.849
So, I'm also, you know, in, in this podcast I'm championing something which may exist, may not exist, uh, which I try to call wildfire engineering.

00:22:25.849 --> 00:22:31.800
And I, I believe it's a job that people will do and they will get paid for its solid money in the future.

00:22:31.800 --> 00:22:36.601
You know, helping communities, uh, protect themselves against hazards like that.

00:22:36.601 --> 00:22:39.601
Therefore, I'm, I'm very open to, you know, gather all.

00:22:39.601 --> 00:22:43.411
That there is on the edge of the research today.

00:22:43.411 --> 00:22:53.371
You know, so the people who are training to become wildfire engineers, whether they know or not, they will be ones, uh, I, I hope they, they, they are best armed for that.

00:22:53.371 --> 00:22:57.605
So continuing with that thought, how do you gather data for that?

00:22:57.605 --> 00:23:03.036
Like, what are you seeking to improve your models, your predictions in that term?

00:23:03.036 --> 00:23:04.415
What kind of data you are looking.

00:23:04.821 --> 00:23:11.403
So this project in particular, I always try to gather mixed methods data.

00:23:11.403 --> 00:23:17.883
So the first step I take, and this is what I did for this project that we've just been talking about, are.

00:23:17.883 --> 00:23:20.643
Qualitative in-depth interviews.

00:23:20.643 --> 00:23:30.363
And so we did interviews with about 30 people, um, for, from each of the three communities that we worked in in Australia.

00:23:30.363 --> 00:23:34.413
And they're amazing people and they gave of their time and told their stories.

00:23:34.413 --> 00:23:40.844
and so these are fire survivors, And, they sat down and talked with my team.

00:23:40.844 --> 00:23:42.344
So it's, um.

00:23:42.344 --> 00:23:44.443
I'm gonna just name them.

00:23:44.443 --> 00:23:46.273
Rosie Morrison is in my team.

00:23:46.273 --> 00:23:47.144
So Dr.

00:23:47.144 --> 00:23:47.864
Uh, Dr.

00:23:47.864 --> 00:23:49.784
Rosie Morrison, uh, Dr.

00:23:49.784 --> 00:23:52.213
Fatima Ru hsa and, uh, Dr.

00:23:52.213 --> 00:23:53.023
Tegan Larin.

00:23:53.023 --> 00:23:59.683
And so they've been, um, involved as research fellows doing these interviews across different communities.

00:23:59.683 --> 00:24:05.233
Tegan has primarily been working in Maui with me after the Maui fires.

00:24:05.233 --> 00:24:11.594
But this project in particular that I've been speaking with you about, um, on travel behavior, has been in Australia.

00:24:11.594 --> 00:24:19.483
And so we looked at, we did interviews in two locations in Western Australia, and then one location in Victoria, which is my home state.

00:24:19.483 --> 00:24:20.233
And.

00:24:20.233 --> 00:24:24.538
That was really just trying to understand, talking with people.

00:24:24.538 --> 00:24:25.979
Please tell us your story.

00:24:25.979 --> 00:24:33.058
Tell us what happened from the first moment that you thought maybe there was a fire going on all the way to when you reached safety.

00:24:33.058 --> 00:24:38.098
And then beyond that, people also talked to us about their recovery and how it was going.

00:24:38.098 --> 00:24:45.670
And, and we just listened and asked, probing questions or follow up questions with what, uh, whatever you, you decide to call them.

00:24:45.670 --> 00:24:47.380
And it was more about.

00:24:47.380 --> 00:24:50.140
Why, um, tell me more about that.

00:24:50.140 --> 00:24:53.740
Please help me to, you know, understand what your thought process was.

00:24:53.740 --> 00:25:01.480
And that is a, is an important step in sort of understanding what are all the reasons why people do what they do.

00:25:01.480 --> 00:25:12.460
And then now we're working on a quantitative survey dive a bit further into the factors that influence people's choices of destinations.

00:25:12.460 --> 00:25:14.440
So do they pick hotel motel?

00:25:14.440 --> 00:25:15.640
Do they go to a friend's house?

00:25:15.640 --> 00:25:18.220
Do they go in an an evacuation shelter?

00:25:18.220 --> 00:25:34.047
Because we're trying to better understand what are the reasons that people make these choices so that we can help communities, better plan for where people will go for safety and maybe how many evacuation shelters they will need.

00:25:34.047 --> 00:25:36.477
What should be the capacity of these shelters?

00:25:36.477 --> 00:25:39.567
Um, and then where people are going otherwise 'cause.

00:25:39.567 --> 00:25:50.547
In our sample, people were traveling upwards of more than a hundred Ks sometimes to go to family and friends, especially if they didn't have a very strong social network.

00:25:50.547 --> 00:25:56.846
Some people were saying, look, Erica, I went all the way to this city because I didn't know anyone else around me.

00:25:56.846 --> 00:26:01.136
And lastly, we do some VR experiments.

00:26:01.136 --> 00:26:15.289
And so one of my PhD students, uh, GHA Atta, she is working on, simulations, driving simulations, using VR to understand how people make routing decisions.

00:26:15.289 --> 00:26:29.660
So if there's, um, particular routes, does the timing of the route, the, the distance of the route, the evacuation cues that they may be exposed to and traffic affect where people travel on the roadways.

00:26:29.660 --> 00:26:32.000
So we're going at it from multiple

00:26:32.342 --> 00:26:42.027
Uh, and how about like, large scale data, like perhaps just the Google, uh, maps, data on traffic congestion, maybe GPS tracking of people.

00:26:42.027 --> 00:26:45.896
Like is, is this something that, that, that's, uh, that you're looking for?

00:26:46.430 --> 00:26:47.539
Absolutely.

00:26:47.539 --> 00:26:52.272
So I've have, great colleagues at the University of Florida.

00:26:52.272 --> 00:27:05.923
Xiao is, uh, is one and working quite a bit with Reno Love Reg as well as part of this team to look at GPS data and how people move on the roadways during fire events.

00:27:05.923 --> 00:27:16.415
GPS data is amazing because it doesn't, rely on people's memory of where they went, whereas obviously we're asking people to try to map their routes.

00:27:16.415 --> 00:27:22.280
The, Difficulty sometimes with GPS data though, is we don't know anything about the person.

00:27:22.280 --> 00:27:26.931
So we know aggregate trends of where people are traveling.

00:27:26.931 --> 00:27:33.441
We have done amazing, Chile's team has done amazing work also with Tom Cova, I should mention, at University of Utah

00:27:33.557 --> 00:27:33.778
Hmm.

00:27:33.980 --> 00:27:44.228
overlaying property location over the maps that we have to understand that people went to residential areas versus maybe a hotel motel.

00:27:44.228 --> 00:27:46.928
So we've, we've done a lot with GPS.

00:27:46.928 --> 00:27:50.917
We've tried to do understand pe where people go for destinations.

00:27:50.917 --> 00:28:00.067
We've done some work understanding evacuation decisions and departure times using, uh, GPS data as well as more recently routing.

00:28:00.067 --> 00:28:01.837
Um, so that's pretty amazing.

00:28:01.837 --> 00:28:13.020
And then the other work that we have been doing with the wound team, so that's led out of, Lund University and the NFPA, uh, research Foundation, Amanda Kimball.

00:28:13.020 --> 00:28:14.820
And then, um, Enrico Roki.

00:28:14.820 --> 00:28:16.969
Trying to improve wound.

00:28:16.969 --> 00:28:20.929
We've looked at trying to mine social media data as well.

00:28:20.929 --> 00:28:27.528
So of course when Twitter was Twitter, um, and not x, it was a little bit easier to gather that information.

00:28:27.528 --> 00:28:40.344
But what we were looking at is could we understand, how people were moving around, how they made their decision to evacuate, maybe where they ended up going for safety, using social media data.

00:28:40.344 --> 00:28:43.013
Lastly, traffic count.

00:28:43.013 --> 00:28:46.644
So there's some work also with Enrico Roki and others.

00:28:46.644 --> 00:29:10.938
He had a PhD student that just finished Arthur Rohar, who had done a lot of work, um, with one of my PhD students, uh, Nima, Jan, Fahan, um, who had done a lot of work looking at traffic count ca traffic, uh, data collected by, the state of California to understand just traffic flows on highways during fires.

00:29:10.938 --> 00:29:16.698
And they found that traffic does move differently during evacuations.

00:29:16.698 --> 00:29:25.458
A bit slower actually, than it would even in, um, congested ingested scenarios than it would in non-emergency situations.

00:29:25.458 --> 00:29:35.712
So we have used a lot of different data trying to come at this problem from a lot of different angles and, um, I think we're getting somewhere and understanding what's going on.

00:29:36.086 --> 00:29:37.105
Oh, this is great.

00:29:37.105 --> 00:29:50.557
Like the large scale data shows you the statistical landscape of, of those decisions and the, the background traffic, as you said, and in depth reviews give you like why people are, are doing the stuff they're doing.

00:29:50.557 --> 00:30:03.336
So I guess it's about finding, nailing the balance between the two and finding the, the links, in regards of, of how people behave and again, my experience with buildings, how, how much people are.

00:30:03.336 --> 00:30:08.041
Prone to nudging in those, uh, evacuation scenarios.

00:30:08.041 --> 00:30:11.816
Like how much drivable is that behavior?

00:30:12.172 --> 00:30:14.717
So that is a, a really good question.

00:30:14.717 --> 00:30:24.166
Uh, we have, when we've been doing interviews, I'll just take Australia for example, and there are definitely people who will stay no matter

00:30:24.229 --> 00:30:24.441
Mm-hmm.

00:30:24.880 --> 00:30:31.965
And we have talked to them about their plans whether or not what they plan to do is what they actually decided to do.

00:30:31.965 --> 00:30:34.516
Um, a lot of people do have triggers.

00:30:34.516 --> 00:30:39.777
Um, so if the fire crosses the ridge or something like that, then we know we need to

00:30:39.900 --> 00:30:40.319
Mm-hmm.

00:30:41.126 --> 00:30:49.767
Um, but for the most part, I think people who will stay have done a lot of the work to do that.

00:30:49.767 --> 00:30:53.156
They've got water tanks, they've got hoses, they've got the protective gear.

00:30:53.156 --> 00:30:54.116
They know what to do.

00:30:54.116 --> 00:30:56.787
They've gone through CFA training to do that.

00:30:56.787 --> 00:31:15.241
but I think what I'm most worried about, to be honest, is people living in, areas that haven't had fire events in the past that are in bushfire prone areas and who, who may not understand their risk.

00:31:15.241 --> 00:31:18.422
And so some of the work I've been doing also.

00:31:18.422 --> 00:31:23.942
Is thinking about how could we help people?

00:31:23.942 --> 00:31:37.481
Let's take suburbs for example, people in suburbs who are not who, who maybe have not had a fire but are actually at fire risk help, help them understand their risk prior to fire season.

00:31:37.481 --> 00:31:40.061
And do some of that nudging that you were talking about.

00:31:40.061 --> 00:31:56.032
And so I'm a big fan of helping people understand in safe and ethical ways, their risks, maybe by virtual reality, um, simulations to really understand like, look, these types of fires are likely to happen here.

00:31:56.032 --> 00:32:00.292
And this is, these are the consequences that could happen if you don't act.

00:32:00.292 --> 00:32:05.843
Um, if you don't leave, this is what could, could be, you know, the scenario that.

00:32:05.843 --> 00:32:08.137
That is quite dangerous to you?

00:32:08.137 --> 00:32:19.553
I think that we could do and help people, especially in areas that haven't had a fire, understand their risk and do a little bit of that safe and ethical nudging you were talking about.

00:32:19.769 --> 00:32:21.309
but, but that's the preparedness phase.

00:32:21.309 --> 00:32:35.374
I, I was thinking more about the exact phase that you're looking into, like the behavior when they're like kind, kind of wayfinding, you know, if you had emergency mode in Google Maps, that would optimize your pathway if you had.

00:32:35.655 --> 00:32:35.875
Oh.

00:32:36.294 --> 00:32:43.884
Like a FEMA app that would like tell you what's the best place, where's your nearest shelter, where is the fire line?

00:32:43.884 --> 00:32:45.413
And, and stuff like that.

00:32:45.413 --> 00:32:56.967
I, I feel those are the tools that, you know, in this modern mindset of, innovating software, uh, so fast, I, I think this is something that could, could, could come up.

00:32:57.444 --> 00:32:57.805
I do.

00:32:57.805 --> 00:32:58.644
I agree with you.

00:32:58.644 --> 00:33:04.734
So we do know that people that we've spoken with actually are looking for evacuation routes.

00:33:04.734 --> 00:33:09.894
It's not something that agencies often provide because things are changing so quickly.

00:33:09.894 --> 00:33:23.920
But I do think with technology, We could eventually, maybe in the short term, get there and provide people with the safe, eh, at the moment path for evacuation because people are looking for that information.

00:33:23.920 --> 00:33:31.029
also working on a project I haven't talked about, we're not finished yet on bushfire predictions.

00:33:31.029 --> 00:33:33.670
And so these are maps that go out.

00:33:33.670 --> 00:33:40.959
They went out in the large, 20 19, 20 20 black summer fires is what they're called here.

00:33:40.959 --> 00:33:45.252
and that was it was a very long and dangerous fire season here.

00:33:45.252 --> 00:33:52.211
and a lot of different areas were threatened and for the first time agencies were releasing maps.

00:33:52.211 --> 00:34:00.580
That would help predict or, or provide, um, worst case predictions for the fire the following day.

00:34:00.580 --> 00:34:05.260
And that actually does help, um, people to understand their risk.

00:34:05.260 --> 00:34:10.510
And this is like in the moment these fires are happening, they released a prediction the night before.

00:34:10.510 --> 00:34:13.090
Um, it goes out to the public.

00:34:13.090 --> 00:34:29.987
And we have done a lot of, uh, interviews and surveys on these types of maps to try to understand what is the most optimal design to get people to pay attention and to leave, if that's what they want people to do.

00:34:29.987 --> 00:34:47.282
And so we've been working really hard on the map design element, but I do think that these are quite powerful products the fire agencies have been ex more excited about using that really do help people to understand like, oh, this could be a really bad day tomorrow and help people to leave early.

00:34:47.918 --> 00:34:55.239
Do you think this could have a significant impact on that pre evacuation time that we're discussing?

00:34:56.230 --> 00:34:57.340
I do, I do.

00:34:57.340 --> 00:35:10.210
I think that if they reach people in time think people have to understand that, look, this might say worse case, but we need to help people to understand uncertainty.

00:35:10.210 --> 00:35:10.871
That's sort of

00:35:10.934 --> 00:35:11.153
Hmm.

00:35:11.440 --> 00:35:12.221
part here.

00:35:12.221 --> 00:35:24.971
Um, but I do, I do think, and from what we've found in some of our studies so far, that this can help to reduce the delay that people will take if they feel like they're at risk.

00:35:24.971 --> 00:35:35.302
If they're really connecting with this map, and if they trust the source, yeah, I, I think it, it could really help people to get out of town earlier than they would have otherwise.

00:35:35.400 --> 00:35:43.469
But at the same time, you cannot tell them to evacuate in the conditions, which they probably should not because they will never use the app again.

00:35:43.469 --> 00:35:43.800
Right.

00:35:43.987 --> 00:35:45.157
it's a, it's a balance.

00:35:45.157 --> 00:35:55.117
We've been talking about this a lot in emergency communication space, and I think there certainly is a tendency for agencies anywhere.

00:35:55.117 --> 00:36:11.213
I'm not just seeking, speaking about Australia, um, worried about if they get the predictions wrong or if something happens, um, that, people maybe did, you know, maybe the prediction said they should do something, they left and it didn't come to fruition.

00:36:11.213 --> 00:36:14.543
the opposite could happen as you suggest.

00:36:14.543 --> 00:36:17.369
but I think the important piece that.

00:36:17.369 --> 00:36:24.199
we could do more is the debrief after an event happens, bringing people together.

00:36:24.199 --> 00:36:38.989
There's actually been research on this that Jeanette Sutton has done out of the university at Albany, um, that shows these debriefing sessions help to increase trust help people to feel like they can do that.

00:36:38.989 --> 00:36:41.150
That they're not gonna just do nothing next time.

00:36:41.150 --> 00:36:47.300
And these debriefs help agencies to explain actually everything that happened exactly as it should have.

00:36:47.300 --> 00:36:53.070
this, this is how our products work and this is why the fire didn't eventuate in the way we thought it would.

00:36:53.070 --> 00:36:55.231
And it actually is quite a powerful tool.

00:36:55.659 --> 00:36:58.478
Do you know how they handle this in tsunami space?

00:36:58.478 --> 00:36:59.918
I'm, I'm from Poland.

00:36:59.918 --> 00:37:03.039
I just had an, I just had an earthquake episode in the podcast.

00:37:03.039 --> 00:37:05.409
I have, like, I don't have earthquakes.

00:37:05.409 --> 00:37:10.298
I have no idea of earthquakes and my knowledge about tsunamis is pretty much the same.

00:37:10.298 --> 00:37:16.652
So, uh, but, but I imagine they must manage that because, I hear about those tsunami warnings.

00:37:16.652 --> 00:37:19.621
Like there's an, there's been an earthquake middle of Pacific Ocean.

00:37:19.621 --> 00:37:23.251
Their tsunami warning has been issued in this, this, and this location.

00:37:23.643 --> 00:37:24.934
it is a good question.

00:37:24.934 --> 00:37:31.713
So they, I think every agency in, in the countries are different, and I don't know how each country

00:37:31.742 --> 00:37:32.032
Okay.

00:37:32.373 --> 00:37:44.284
that, because you're absolutely right for earthquakes in particular, when people get a tsunami warning, it, it's a, it's a tough one because they don't exactly know if a tsunami will eventuate or

00:37:44.621 --> 00:37:44.842
Mm.

00:37:44.914 --> 00:37:46.443
they want people to be safe.

00:37:46.443 --> 00:37:53.853
I actually don't know if a lot, you know, which countries are excelling on the debriefing aspect of that.

00:37:53.853 --> 00:37:55.054
I would imagine.

00:37:55.054 --> 00:37:56.043
I'm just gonna pause it.

00:37:56.043 --> 00:37:56.974
I'm gonna put it out there.

00:37:56.974 --> 00:37:59.643
It probably isn't the norm.

00:37:59.643 --> 00:38:25.594
To do that because resources in fire, emergency agencies are not always flush and they have competing prior, like a lot of times, especially in this country here, something will happen and then another thing will happen, and then I think agencies are always trying to catch up and always responding, and so it's hard to go back to community and have that conversation.

00:38:25.594 --> 00:38:26.434
I'm not saying it's never

00:38:26.831 --> 00:38:27.161
Mm.

00:38:27.304 --> 00:38:33.273
but I'm saying it's difficult to do because of the number of events that are always happening in locations.

00:38:33.791 --> 00:38:35.387
Perhaps I should rephrase the question.

00:38:35.387 --> 00:38:48.342
I'm wondering how much you should arm the people with information that allows them to take sovereign decision based on the information itself.

00:38:48.342 --> 00:38:52.677
How much you should tell them, you know, this is the moment you should evacuate.

00:38:52.677 --> 00:38:54.867
Like, this is the time you should, you should go.

00:38:54.867 --> 00:39:00.907
And kind of, you know, giving them decision right out like you need to go.

00:39:00.907 --> 00:39:03.996
But then, you know, the, the trust factor.

00:39:03.996 --> 00:39:14.215
If, if, if it wasn't urgent, if, if you tell them to evacuate, uh, every week after three times, they're not gonna evacuate for the full time.

00:39:14.215 --> 00:39:30.715
So I, I find this is, this is an interesting because we probably would love to provide, you know, better information to people, better guidance and act on the factors that you're researching at the other end.

00:39:30.715 --> 00:39:32.155
We don't want to break it.

00:39:32.155 --> 00:39:37.619
And I'm speaking now from my experience with, uh, false alarm and fire detection systems, you know.

00:39:37.682 --> 00:39:38.822
yes, yes, yes.

00:39:39.190 --> 00:39:45.610
We can detect the slightest amount of smoke, but we don't necessarily want to evacuate the 10,000 people.

00:39:45.610 --> 00:39:46.360
Right.

00:39:46.583 --> 00:39:46.764
Yep.

00:39:46.764 --> 00:39:47.994
Where's the threshold?

00:39:47.994 --> 00:39:52.878
I'm, I am gonna err on the side of tell them to go.

00:39:52.878 --> 00:39:54.137
really

00:39:54.335 --> 00:39:54.625
Yeah.

00:39:54.737 --> 00:39:55.518
I'm gonna be.

00:39:55.518 --> 00:40:01.550
And I, you know, maybe may, maybe when you talk to me in another five years, I will say, ah.

00:40:01.550 --> 00:40:02.889
I shouldn't have

00:40:02.996 --> 00:40:03.286
yeah.

00:40:03.550 --> 00:40:06.268
No, but I, I, I think it's better to be safe.

00:40:06.268 --> 00:40:19.617
And I, and from the Maui work, which hopefully the report will be coming out soon, um, where they had a loss of power and a loss of service, self services and they weren't able to get the word out.

00:40:19.617 --> 00:40:22.288
People were waiting for official word to go.

00:40:22.288 --> 00:40:22.708
People

00:40:22.795 --> 00:40:23.085
Okay.

00:40:23.608 --> 00:40:24.748
for that information.

00:40:24.748 --> 00:40:24.927
They

00:40:25.155 --> 00:40:25.445
Okay.

00:40:25.768 --> 00:40:28.047
and they want the information.

00:40:28.047 --> 00:40:29.253
go.

00:40:29.253 --> 00:40:34.577
And, this is especially the case because they didn't have a lot of experience with fires.

00:40:34.577 --> 00:40:39.708
They had a lot of experience with cyclones or hurricanes and tsunami.

00:40:39.708 --> 00:40:50.090
and so it's even more important to tell people to go, especially in areas where they're not as experienced and they're likely to be waiting for official word.

00:40:50.090 --> 00:40:54.559
So I'm really airing on both on that side.

00:40:54.559 --> 00:40:57.559
But I do think it's helpful to have multiple products.

00:40:57.559 --> 00:41:03.170
So those predictions that go out the night before, those are hopefully gonna help people who will leave early.

00:41:03.170 --> 00:41:03.769
Okay.

00:41:03.769 --> 00:41:05.719
Then we know not everyone's gonna leave early.

00:41:05.719 --> 00:41:13.219
So then you give the message to go now, leave immediately, and then hopefully you get most of the other people who are going to leave.

00:41:13.219 --> 00:41:18.949
And then in Australia, they've got a too late to leave message so they.

00:41:18.949 --> 00:41:23.849
They will, fire agencies will say, um, it is too late to leave.

00:41:23.849 --> 00:41:28.650
You have to bunker down now because it's, it's no longer safe for you to be on the roadways.

00:41:28.650 --> 00:41:34.110
And so then they help people to understand like, okay, all right, now, too late to leave.

00:41:34.110 --> 00:41:37.409
I've gotta do what I need to do and defend in place.

00:41:37.568 --> 00:41:38.557
That's horrible.

00:41:38.557 --> 00:41:45.518
I mean, that, that's such a, I mean, I'm not saying that the information, I mean, it's a horrible situation to find yourself in.

00:41:45.867 --> 00:41:51.838
It is, but it's a very helpful message because they don't want people getting on the roadways and,

00:41:51.951 --> 00:41:52.010
Yeah.

00:41:52.318 --> 00:41:52.918
themselves.

00:41:52.971 --> 00:41:53.150
mean, yeah.

00:41:53.248 --> 00:42:02.465
And the fire agencies really help people to understand if you are in a place where you actually can't leave, these are the things that you need to do.

00:42:02.465 --> 00:42:09.186
So there's a lot of really good preparedness information out there for people to understand, if I'm in this situation, this is what I do.

00:42:09.186 --> 00:42:10.775
If I'm in this situation, this is what I

00:42:10.887 --> 00:42:11.367
Hmm.

00:42:12.155 --> 00:42:17.525
So it is very good information, but I completely understand it it's never a good situation to

00:42:17.652 --> 00:42:18.163
It's quite,

00:42:18.306 --> 00:42:18.735
I, I

00:42:18.913 --> 00:42:20.382
can imagine it's, it's quite traumatic.

00:42:20.382 --> 00:42:43.461
I mean, in, in this, in this kind of informing people, there's also like this dynamic that, if you have a wildfire developing like, uh, 20 miles from where you live, and it's like moving forward and it's getting closer and closed and you observe how it's getting closer and then you get information, you should go, that's a completely different story than compared to like campfire.

00:42:43.461 --> 00:42:47.181
It's, it's, it starts, bam, three hours later, 8:00 AM in the morning.

00:42:47.181 --> 00:42:48.561
You need to go now, man.

00:42:48.561 --> 00:42:50.902
It, it's, it's, it's already almost late.

00:42:50.902 --> 00:42:53.572
Like those are completely different dynamics as well.

00:42:53.668 --> 00:42:54.989
I completely agree with you.

00:42:54.989 --> 00:42:58.648
And that's why NIST has been working and, and others have been working.

00:42:58.648 --> 00:43:10.438
Tom Cova has been working on understanding dire fire situations, DIRE, dire fire situations where they have very little time to leave and then.

00:43:10.438 --> 00:43:22.739
In these situations, should we be thinking about in fire affected areas, having a safe place for people to go, a safe, protected place as a la place of last resort?

00:43:22.739 --> 00:43:26.909
Um, you can't leave you, you know, it's come on too fast.

00:43:26.909 --> 00:43:32.068
Or maybe you are a person who needed extra time to leave and there isn't that time.

00:43:32.068 --> 00:43:38.688
And so that's why thinking about places of last resort for these types of scenarios are so, so important.

00:43:38.688 --> 00:43:41.688
I think we have to keep this in mind.

00:43:41.688 --> 00:43:46.789
You bring up such a good point about the differences in fire dynamics, right?

00:43:46.789 --> 00:43:50.628
We've got fires that have been, as you said, been rolling along the cliffs and,

00:43:51.306 --> 00:43:51.880
you see them?

00:43:51.880 --> 00:43:53.710
They are two hills away now.

00:43:53.710 --> 00:43:55.061
They're one hill away.

00:43:55.061 --> 00:43:57.101
You seen the smoke, you seen the flames?

00:43:57.101 --> 00:43:57.550
Yeah.

00:43:57.588 --> 00:43:57.918
Yep.

00:43:57.918 --> 00:44:03.108
And that's what happened in the chimney Tops fire, the very first fire that I studied in Tennessee.

00:44:03.108 --> 00:44:09.554
And that's some of the, Studies that we look at, people said, yeah, I saw smoke.

00:44:09.554 --> 00:44:17.974
And a lot of times we say, oh, if you see environmental cues, that is very much meaning that your risk is high and that you leave.

00:44:17.974 --> 00:44:27.755
But when you've been seeing smoke in the horizon for days and days and days and days, that is not an environmental cue that is gonna increase your risk and cause you to go.

00:44:27.755 --> 00:44:38.710
And so understanding the length of fires, um, the, the cues that people have been seeing over time, or as you said, all of a sudden it ramps up and there is no choice.

00:44:38.710 --> 00:44:43.119
Completely different scenarios and meaning completely different behavior by people.

00:44:43.773 --> 00:44:49.443
I'll, I'll put another twist on that I was researching and also find, found another interesting paper.

00:44:49.443 --> 00:44:55.833
I mean, it's not very hard because there's like a ton of interesting papers you have, but, uh, but I found this one.

00:44:55.833 --> 00:44:58.894
Um, how about people who are not the citizens?

00:44:58.894 --> 00:45:12.628
How about people who are there, like tourists or, you know, they, they, they're just visiting or they're just passing by and they're, they're absolutely not, familiar with the local fire season or, or fire behaviors.

00:45:12.628 --> 00:45:15.463
Um, how, how different is, is, is it in that case?

00:45:16.213 --> 00:45:23.112
This is such an important question and something that, um, Enrico and Lund has been working on.

00:45:23.112 --> 00:45:35.443
And I also a new study that we just finished, um, looking at a fire that was, that happened in a touristy area in December, 2024.

00:45:35.443 --> 00:45:38.052
And so hopefully that report will come out soon.

00:45:38.052 --> 00:45:40.032
But what we found just.

00:45:40.032 --> 00:45:48.163
Kind of an overarching look at our findings is that we did talk to visitors.

00:45:48.163 --> 00:45:49.333
It was a smaller group.

00:45:49.333 --> 00:45:50.233
It was a smaller group.

00:45:50.233 --> 00:45:52.242
So we wish we would've been able to speak with more.

00:45:52.242 --> 00:45:57.793
And there were visitors from inside of the state, inside of Australia and international.

00:45:57.793 --> 00:46:08.811
And we saw very much, uh, differences in how people were seeking information or even understanding where to go for information depending upon where they were.

00:46:08.811 --> 00:46:18.951
So if they came from my state and they went, you know, a couple of hours up to the Grampians, which is where this fire happened, they had the app on their phone.

00:46:18.951 --> 00:46:21.440
They knew the fire risk and they were set.

00:46:21.440 --> 00:46:23.541
And a lot of times they didn't even travel.

00:46:23.541 --> 00:46:27.380
They saw that it was dangerous and they decided not to go.

00:46:27.380 --> 00:46:35.150
Whereas if you have international, they, they're not going to be able to have known what app to download, nowhere to go for information.

00:46:35.150 --> 00:46:37.795
And what we were interested in is were the.

00:46:37.795 --> 00:46:43.561
Tourists, like were there facilities where they're staying, communicating with them.

00:46:43.561 --> 00:46:45.150
And in some cases they were getting

00:46:45.179 --> 00:46:45.599
Mm-hmm.

00:46:45.840 --> 00:46:47.641
from their accommodation provider.

00:46:47.641 --> 00:46:55.320
In some cases, maybe not as much because the accommodation provider was also trying to protect their property and themselves.

00:46:55.320 --> 00:47:02.251
And so it gets to be quite complicated when we talk about tourists, how they seek information and what they decide to do.

00:47:02.251 --> 00:47:09.840
And I think we need much more research on how to communicate with international visitors.

00:47:09.840 --> 00:47:13.141
It was actually a very cool, so we have, sorry if I'm going off

00:47:13.219 --> 00:47:13.934
No, please go.

00:47:14.010 --> 00:47:17.896
but we have a, a Natural Hazards Research Australia, which is a.

00:47:17.896 --> 00:47:25.797
National research funding body here in Australia, and they, every year have a disaster challenge.

00:47:25.797 --> 00:47:29.728
And one of them was how to communicate with hard to reach groups

00:47:29.795 --> 00:47:30.186
Mm-hmm.

00:47:30.297 --> 00:47:32.306
and, it's normally like teams.

00:47:32.306 --> 00:47:35.755
It could be students, it could be PhD students, it could be research fellows.

00:47:35.755 --> 00:47:49.351
And the team that won had really interesting idea that when you are out and about, like say you're at the beach and you're trying to get on the wifi, before you get onto wifi, you have a video that you see.

00:47:49.351 --> 00:47:55.516
like you've gotta watch this video for 30 seconds or 45 seconds before you can get onto wifi.

00:47:55.516 --> 00:48:03.407
And that video tells you, here's the app to download, here's where you get information, here's what the fire risk is in your area.

00:48:03.407 --> 00:48:10.936
I don't know if they've moved that forward at all, but it's such, it's such a good way to, you've got a captured audience.

00:48:10.936 --> 00:48:13.456
They need the internet for what they're doing.

00:48:13.456 --> 00:48:18.244
They're international, so they're not gonna be on, you know, data, or they may not be on data.

00:48:18.244 --> 00:48:24.994
And then you have an opportunity to tell them a bit more about their fire risk and how to get information about what to do in a fire.

00:48:24.994 --> 00:48:28.324
So we've got a lot of work to do, I think, for tourists.

00:48:28.496 --> 00:48:35.630
I wonder how effective would be the, your usual means of letting people know, like, you know, sirens and, and

00:48:35.773 --> 00:48:38.135
Ah, you're asking some great questions

00:48:38.217 --> 00:48:42.202
did my research 40 episodes in.

00:48:42.501 --> 00:48:43.550
That's professional.

00:48:43.550 --> 00:48:45.050
I know you are.

00:48:45.050 --> 00:48:47.931
I would never, never say otherwise.

00:48:47.931 --> 00:48:56.907
So sirens, oh, such a good question and So sirens are extremely helpful, but they've got their limitations.

00:48:56.907 --> 00:49:00.987
So sirens are only meant to warn people outdoors.

00:49:00.987 --> 00:49:05.728
Um, and we can't rely on them in indoors.

00:49:05.728 --> 00:49:07.677
Um, they're not what they're meant to do.

00:49:07.677 --> 00:49:18.987
And depending upon the design of the siren, um, sometimes you can have information provided along with the sound, but sometimes that's quite difficult because the siren could be like a 360 degree.

00:49:18.987 --> 00:49:24.507
It could be one that turns and you only hear it like when it's in facing in your direction.

00:49:24.507 --> 00:49:30.184
But sirens need to be accompanied be accompanied with information.

00:49:30.184 --> 00:49:36.094
Something telling people this is what the siren is about and this is what you need to do.

00:49:36.094 --> 00:49:53.014
Sirens are only meant to alert us, and they're only effective if people know where to go for further information or if you give information as part of that siren system, because that's the warning piece that we absolutely need, that the siren isn't going to be enough.

00:49:53.014 --> 00:50:07.565
And there's been a lot of, sort of back and forth about whether, um, you know, the sirens exec, for example, should have sounded, uh, in the Maui fire, uh, that happened in 2023, uh, in Lena.

00:50:07.565 --> 00:50:27.494
And, I think the thing that's important to recognize is it's not just good enough to just sound it and make sure that, well, the information is vital to have at the same time, it's also important to train and educate the population on the siren for specific types of events.

00:50:27.494 --> 00:50:43.784
So if it's only been used for tsunami, or it's only been used for cyclones, but then for this event you wanna use it for fire, that's a difficult decision because the community wouldn't have been trained to associate sirens with fire.

00:50:43.784 --> 00:50:51.853
Um, but more important above all of that is that information needs to be, come along to, needs to come along with the siren

00:50:52.221 --> 00:50:57.981
they probably would also have a challenge in distinguishing like, if there's like three different sounds.

00:50:57.981 --> 00:51:01.911
Like it is not easy to interpret if it's just the sound of siren.

00:51:01.911 --> 00:51:06.831
And if I put myself in the position of a tourist, you know, I would just look at locals, what they are doing.

00:51:06.831 --> 00:51:10.820
So it's, it's really key importance to train the local population.

00:51:10.820 --> 00:51:13.764
And that also kind of helps with the, tourists.

00:51:14.021 --> 00:51:14.561
For sure.

00:51:14.561 --> 00:51:15.371
Absolutely.

00:51:15.371 --> 00:51:16.152
100%.

00:51:16.152 --> 00:51:21.012
But sirens is a good way to get people alerted that something is going

00:51:21.264 --> 00:51:21.344
Hmm.

00:51:21.822 --> 00:51:30.612
and then hopefully you get them to turn on the radio or turn on the tv, but you have to make sure that the information's there or people tune it off.

00:51:30.612 --> 00:51:34.117
They're not saying anything about it on the radio, so it must not be

00:51:34.215 --> 00:51:37.846
Uh, I have, I have one really hilarious siren story.

00:51:37.846 --> 00:51:41.724
So I had, a student, uh, Diego from Spain.

00:51:41.724 --> 00:51:47.543
He, he won an SFP research grant, and he came to my office to do tunneling research.

00:51:47.543 --> 00:51:58.284
And, and he came to Warsaw and he chose a hotel in the middle downtown Warsaw next to the big Soviet building in very middle of Warsaw.

00:51:58.284 --> 00:52:00.083
He came to Poland.

00:52:00.083 --> 00:52:02.213
Uh, he, he, he met me in the office.

00:52:02.213 --> 00:52:03.083
He went to the hotel.

00:52:03.083 --> 00:52:07.853
I forgot to tell him like it was the 1st of August.

00:52:07.853 --> 00:52:11.664
1st of August is the anniversary of Warsaw Uprising.

00:52:11.664 --> 00:52:17.393
So it is a very big deal in Warsaw, like you can Google in YouTube, the city that stops for a minute.

00:52:17.393 --> 00:52:21.266
5:00 PM Everyone in the city stops.

00:52:21.266 --> 00:52:26.126
The whole city goes dead, sirens go everywhere.

00:52:26.126 --> 00:52:36.166
And in the location, very front of his hotel, there's probably gathering of like a hundred thousand people with, you know, flowers and everything shouting.

00:52:36.166 --> 00:52:43.005
Like there's basically one sea of fire and smoke at 5:00 PM accompanied by sirens.

00:52:43.005 --> 00:52:49.353
And he's there at 20 floor of that hotel, absolutely unaware of that fact.

00:52:49.353 --> 00:52:57.302
And suddenly sirens go off, he looks outside and there's a flood of fire air going out.

00:52:57.639 --> 00:52:58.599
gosh,

00:52:58.922 --> 00:53:01.353
the cultural context is so sometimes

00:53:01.389 --> 00:53:02.349
yes.

00:53:02.349 --> 00:53:04.840
What What did he think it was?

00:53:04.840 --> 00:53:05.829
I'm so curious.

00:53:05.829 --> 00:53:06.579
Did he tell you?

00:53:06.918 --> 00:53:10.233
Like he thought it is, like uprising or something.

00:53:10.233 --> 00:53:10.893
Like, he was

00:53:11.139 --> 00:53:11.230
geez.

00:53:11.282 --> 00:53:16.202
very, very very confused, but there was no shots fired.

00:53:16.202 --> 00:53:18.063
So he, he assumed that it's probably,

00:53:18.074 --> 00:53:18.175
God.

00:53:18.362 --> 00:53:19.592
probably fine.

00:53:19.592 --> 00:53:26.244
I really forgot to tell him like, dude, it's, first of all, 'cause there there's gonna be some interesting things happening.

00:53:26.244 --> 00:53:29.094
I just really, really, really should have told him.

00:53:29.094 --> 00:53:30.295
Oh man.

00:53:30.295 --> 00:53:36.820
But, uh, anyway, uh, to, to finish this off, what kind of gaps do you see?

00:53:36.820 --> 00:53:40.061
Like what's the biggest hurdle for you right now?

00:53:40.061 --> 00:53:52.226
Data models, complexity, like what's the biggest, uh, blockage on the pathway to having this worked out into something that we could practically use.

00:53:52.864 --> 00:54:09.719
I think that, that, that's sort of my next hurdle is to collect more quantifiable data and then work with evacuation modelers to try to translate that, those findings into, into

00:54:10.032 --> 00:54:10.452
Mm-hmm.

00:54:10.829 --> 00:54:15.599
So I wouldn't maybe say it's a gap, but is it is a true future goal.

00:54:15.599 --> 00:54:19.090
I know we can do it, but it takes a bit more data.

00:54:19.090 --> 00:54:28.601
I think, as you also nicely alluded to, like earlier in our, in our talk, that we do need to bring in.

00:54:28.601 --> 00:54:31.871
All, all of the data sets that we can.

00:54:31.871 --> 00:54:40.061
And what I would like to see more of, we've tried to do a little bit of this, but it's difficult is to have multiple data sets for the same

00:54:40.293 --> 00:54:40.713
Mm-hmm.

00:54:41.081 --> 00:54:59.291
So we, we've got the large scale sets and then we've got some interviews or surveys that we can try to put some understanding behind what we're finding because it's a lot easier to quantify the larger sets, but a little bit difficult if we don't know sort of some of the whys behind it.

00:54:59.291 --> 00:55:24.170
So I don't necessarily think that there are big gaps that we can't overcome the moment, but I just think it's bringing the teams together and having the different disciplines talk to one another and the social sciences, talk with the engineers, speak with the computer scientists, and work very well together, and to try to bring in some of these social science findings into the tools.

00:55:24.782 --> 00:55:31.288
Is it hard for you to go from the modeling of like, let's say building scale to a, a community or regional scale?

00:55:31.288 --> 00:55:33.027
Do you use different tools?

00:55:33.027 --> 00:55:35.722
Uh, how, how does this scale shift look like?

00:55:36.201 --> 00:55:36.440
Yeah.

00:55:36.440 --> 00:55:39.231
So it actually, it was, it was quite difficult for me.

00:55:39.231 --> 00:55:44.001
I didn't necessarily think that I could go directly from buildings into wildfires.

00:55:44.001 --> 00:55:47.001
I, I didn't have that thought.

00:55:47.001 --> 00:55:55.632
The first thing I did was met with some of the giants in social science, like Sarah McCaffrey, uh, from the Forest Service recently retired.

00:55:55.632 --> 00:55:57.000
and I.

00:55:57.000 --> 00:56:04.762
I wanted to understand just the lay of the land in wildfires, but the evacuation tools that we use are different.

00:56:04.762 --> 00:56:10.913
So in building fires, right, we're using the building simulation tools in wildfires.

00:56:10.913 --> 00:56:13.621
had to develop some tools.

00:56:13.621 --> 00:56:15.061
Wound is one of them.

00:56:15.061 --> 00:56:19.650
Um, NIST funded, uh, ity tool that brings in.

00:56:19.650 --> 00:56:29.775
It's sort of a modular approach in bringing in the, and I know you've had podcasts on this, bringing in the fire modeling with the traffic modeling, with the pedestrian modeling.

00:56:29.775 --> 00:56:33.795
And so that is new for, for me.

00:56:33.795 --> 00:56:36.795
I'm not a traffic, engineering expert.

00:56:36.795 --> 00:56:40.394
Um, and so I've had to sort of dig into some of those.

00:56:40.394 --> 00:56:51.324
you know, to dig into some of the reports, looking at a lot of non-emergency transport studies in how they collect data on what people's preferences are.

00:56:51.324 --> 00:56:54.864
But also, like you said, the tools are different.

00:56:54.864 --> 00:56:56.724
They've had to be sort of developed.

00:56:56.724 --> 00:57:03.715
There's also a tool, um, in Australia developed by C-S-I-R-O that also does bushfire evacuation.

00:57:03.715 --> 00:57:06.114
So there are tools that exist.

00:57:06.114 --> 00:57:11.425
They've had to sort of bring in the traffic and the pedestrian and the fire, which is unique.

00:57:11.425 --> 00:57:14.106
we don't obviously need the traffic for the building

00:57:14.173 --> 00:57:14.393
Hmm.

00:57:14.585 --> 00:57:17.195
but we do need the pedestrian and the fire stuff.

00:57:17.195 --> 00:57:21.666
So there is an some added comple complexity to that.

00:57:21.666 --> 00:57:24.365
But honestly for me, that's what makes it really interesting.

00:57:24.365 --> 00:57:36.068
I have really enjoyed, uh, diving into a new field and meeting new people and learning, how to do this well and actually try to protect people in bushfire prone areas.

00:57:36.126 --> 00:57:40.806
And, and I think what you're doing is, is not just relevant for bushel wildfires.

00:57:40.806 --> 00:57:53.021
I think at some point a lot of that will come back to the building people because, you know, a human is a human and, and I, I already see a lot of things that could be.

00:57:53.021 --> 00:58:10.777
Incorporated in building evacuation, especially when we talk about, you know, optimizing the evacuation and, uh, minimizing the whatever you would like to minimize hazard risk, downtime, uh, waiting time, whatever your variable would be.

00:58:10.777 --> 00:58:12.451
These are also important.

00:58:12.552 --> 00:58:13.541
I agree.

00:58:13.541 --> 00:58:17.023
And I just wanted to say one thing, you mentioned something really important.

00:58:17.023 --> 00:58:21.793
You said that you think that fire engineers, um, will be working in the wildfire space

00:58:21.896 --> 00:58:22.346
They will.

00:58:22.813 --> 00:58:24.492
And I completely agree with you.

00:58:24.492 --> 00:58:37.782
I feel like that's what the Society of Fire Protection Engineers is working on through the research foundation and their wooey focus and, and helping to understand the role of fire safety and fire protection engineers in the wildfire space.

00:58:37.782 --> 00:58:42.672
And as you know, like as fire protection or fire safety engineers, we learn how to use.

00:58:42.672 --> 00:58:44.972
The evacuation models.

00:58:44.972 --> 00:58:46.983
We, we learn that these tools exist.

00:58:46.983 --> 00:58:53.342
We, we ha you know, we've got to understand how to calculate evacuation from a building.

00:58:53.342 --> 00:59:05.552
And I think that we can those same tools, um, some training on the wildfire tools that exist and really use our knowledge from r set and a set.

00:59:05.552 --> 00:59:18.961
And, and Enrico's over and, and team have already been sort of working and bringing that NRC, university of Greenwich already been bringing a set and r set to the wildfire space with and wa set.

00:59:18.961 --> 00:59:20.911
So this really, really exciting work.

00:59:20.911 --> 00:59:25.981
Imperial has also been involved, so I think there's a big role for fire safety engineers.

00:59:25.981 --> 00:59:26.552
If I could just

00:59:26.724 --> 00:59:28.105
Yeah, abs no, absolutely.

00:59:28.105 --> 00:59:32.364
I hope just don don't, they don't make BA and B Rs for BFIs.

00:59:32.364 --> 00:59:32.940
But,

00:59:33.422 --> 00:59:33.692
Yeah,

00:59:34.135 --> 00:59:34.315
but

00:59:34.351 --> 00:59:38.342
we'll just take the w we'll just, we'll take, we'll take a W set and w set.

00:59:38.349 --> 00:59:46.045
I, I, while I a hundred percent agree with what you just said, I've, I see the same thing from the opposite side.

00:59:46.045 --> 01:00:02.632
I see communities that could use this, and I see communities as clients who have money for this because this type of civil defense stuff is actually something that is fundable at the regional level.

01:00:02.632 --> 01:00:09.112
If I think about European Union landscape, this is something you could get you funding for, like.

01:00:09.112 --> 01:00:16.588
I absolutely see you being a community that believes they are, prone to such a hazard.

01:00:16.588 --> 01:00:18.327
Reach out for EU funding.

01:00:18.327 --> 01:00:20.847
We need to improve our resilience, right?

01:00:20.847 --> 01:00:27.391
A small project, get a hundred thousand euros, hire a fire safety engineer and get something done for yourself.

01:00:27.391 --> 01:00:37.853
I see a commercial pathway with people having pathways to get funds and spend those funds on, on good service, which fire safety engineers could provide in the future.

01:00:37.853 --> 01:00:40.193
I, I really see this from the business perspective.

01:00:40.666 --> 01:00:40.905
Hmm,

01:00:41.003 --> 01:00:41.302
Uh,

01:00:41.356 --> 01:00:41.746
do too.

01:00:41.746 --> 01:00:51.635
I think there are some, um, businesses already that are offering a wildfire, platform, working with communities or working with, uh, third parties, working with communities.

01:00:51.635 --> 01:00:52.806
So I absolutely agree.

01:00:52.806 --> 01:00:55.175
I think it's different depending upon what country

01:00:55.333 --> 01:00:55.753
absolutely.

01:00:55.753 --> 01:00:56.534
Of course, of course.

01:00:56.534 --> 01:00:56.923
Yeah.

01:00:57.036 --> 01:01:00.726
so maybe it's sort of, it will continue to grow.

01:01:01.289 --> 01:01:05.338
But it has to major a little bit, but it's going into that direction.

01:01:05.338 --> 01:01:07.739
It definitely is going that, that direction.

01:01:08.106 --> 01:01:08.496
agree.

01:01:08.518 --> 01:01:17.762
Uh, one thing I I need to say about human behavior in fire community, this community is one of the best integrated parts of fire science.

01:01:17.762 --> 01:01:24.211
And even, you know, this interview you've named, dropped like 30 different people from like 20 different institutions.

01:01:24.211 --> 01:01:25.742
And I, I really appre

01:01:25.960 --> 01:01:26.710
Sorry about

01:01:26.882 --> 01:01:27.751
No, no, no.

01:01:27.751 --> 01:01:28.246
I, I, I,

01:01:28.449 --> 01:01:28.809
people.

01:01:29.036 --> 01:01:35.192
I, I I really am appreciate that because that's the level of integration that most parts of the fire science don't have.

01:01:35.192 --> 01:01:42.632
And if you don't mind on that note, I'll go, I'm gonna drop the fact that, uh, there's a conference happening in Lund, uh, in October.

01:01:42.632 --> 01:01:43.742
Uh.

01:01:43.742 --> 01:01:52.561
Maybe we, we probably should mention that it's six to eighth, October 26th at, uh, campus Helsingborg in Lund.

01:01:52.561 --> 01:01:56.282
I've just learned that it's like 30 minutes away from Lund.

01:01:56.282 --> 01:02:00.061
Um, I hope to visit Enrico, uh, next week in that campus.

01:02:00.061 --> 01:02:07.952
So I've just researched and it's as easy to reach as learned is, and I think it's gonna be packed with some really nice talks.

01:02:07.952 --> 01:02:09.512
I saw you in the organizing committee.

01:02:09.512 --> 01:02:09.931
Can you

01:02:10.059 --> 01:02:10.539
yes.

01:02:10.539 --> 01:02:10.900
I,

01:02:10.922 --> 01:02:12.302
share and advertise a little bit?

01:02:12.681 --> 01:02:14.900
It is gonna be an exciting conference.

01:02:14.900 --> 01:02:18.351
Um, we got a number of really great abstracts.

01:02:18.351 --> 01:02:23.751
I know that, um, Daniel Nielsen and, uh, Enrico Ronchi are going through them now.

01:02:23.751 --> 01:02:30.056
We have, given our scores and so hopefully we'll be hearing soon who the, about the program.

01:02:30.056 --> 01:02:37.797
It should be a really exciting one about buildings and about wildfires and, um, and other, maybe even some safety and security.

01:02:37.797 --> 01:02:45.987
So stay tuned and hopefully you'll, you'll, uh, be able to attend because it's gonna be some exciting stuff and it hasn't happened for us in a while.

01:02:45.987 --> 01:02:56.039
So this is sort of bringing back the conference, has, it's been a little while since we've had the last one, so I'm very excited about it and I, hopefully, hopefully other people will be able to attend.

01:02:56.202 --> 01:02:57.757
It, It, it, looks, uh.

01:02:57.757 --> 01:02:58.942
Solid.

01:02:58.942 --> 01:02:59.811
And there's also,

01:02:59.878 --> 01:03:00.119
Yes,

01:03:00.351 --> 01:03:06.101
and I think KO is also having a workshop related to accessibility in his, ERC grant.

01:03:06.101 --> 01:03:08.081
So that's very interesting.

01:03:08.081 --> 01:03:10.259
next week I'll know more because I'll visit him.

01:03:10.259 --> 01:03:13.829
So, uh, if, if, if I learn more, I'll, I'll share more online.

01:03:13.829 --> 01:03:16.168
And, and for now, uh, Ika, thank you.

01:03:16.168 --> 01:03:28.876
Thank you so much for, for joining me here and, and sharing, uh, the progress on, on those, large scale fire evacuations and, and what happens between when people learn about the, the fire and, uh, evacuate safely.

01:03:29.784 --> 01:03:30.864
Thank you for having me on.

01:03:30.864 --> 01:03:32.454
I've really enjoyed this talk today.

01:03:32.454 --> 01:03:32.994
It's been fun.

01:03:32.994 --> 01:03:33.023
I

01:03:33.510 --> 01:03:34.019
And that's it.

01:03:34.019 --> 01:03:34.800
Thank you for listening.

01:03:34.800 --> 01:03:43.679
As promised, it was not just an interview about the decision making, but also a high level overview of the entire field of human behavior in fires.

01:03:43.679 --> 01:03:47.144
It's such a great field, great people in it.

01:03:47.144 --> 01:03:51.434
Uh, as I sat, I'm visiting Enrico, I just finished visiting him.

01:03:51.434 --> 01:03:55.215
He put weird stuff on my head and made me evacuate in vr.

01:03:55.215 --> 01:03:56.655
It was fantastic.

01:03:56.655 --> 01:04:06.434
And, uh, he, the, I've checked the Helburg campus of Lund University does exist and is the place where the big human behavior in foreign evacuation conference.

01:04:06.434 --> 01:04:10.170
Will happen this year for which all of you are invited.

01:04:10.170 --> 01:04:12.659
And, uh, link is in the show notes.

01:04:12.659 --> 01:04:19.289
As for the episode, I think it's very interesting to investigate what people do after they choose to leave.

01:04:19.289 --> 01:04:29.309
You know, there's something like, if you think about the classical phases of evacuation, uh, pre evacuation, distribution times, you know, the stuff that we will be normally accounting for.

01:04:29.309 --> 01:04:31.920
This is absolutely something I've never been accounting for.

01:04:31.920 --> 01:04:32.849
Like if.

01:04:32.849 --> 01:04:51.757
Someone decided to evacuate, they go in my models, if someone is, uh, over the time they're supposed to evacuate, unless I have an extremely specific task, uh, for the person, like a return for a person that needs help, for example, and I explicitly program that.

01:04:51.757 --> 01:04:54.668
I don't really account for that.

01:04:54.668 --> 01:05:01.027
So interesting to, to see what that background noise can do to your evacuation process.

01:05:01.027 --> 01:05:07.628
And I hope Erica succeeds in developing those models and they become a useful tool for our profession.

01:05:07.628 --> 01:05:09.907
That would be it for today's episode.

01:05:09.907 --> 01:05:14.467
I hope you've enjoyed a little bit of human behavior in fires and the podcast.

01:05:14.467 --> 01:05:18.847
And, uh, come back here next week for more fire science coming your way.

01:05:18.847 --> 01:05:20.887
See you there next Wednesday cheer.

01:05:20.887 --> 01:05:21.427
Bye.