Episode 333: A Digital Sand Table for Wildland Fire with Jackson Yip of Paladin Industries

Episode notes

Wildland crews dig line for days. Then the fire jumps it, and they fall back to the next ridge. Nobody knew if that line would hold. Jackson Yip wants crews to know before they dig.

Hello Smart Firefighting Community!

Welcome to another episode of covering real world innovations via interviews with fire service and technology industry experts that empower YOU to develop your very own Smart Firefighting strategy.

Jackson is a meteorologist and the founder of Paladin Industries . He came up through the fire weather lab at San Jose State. He was in Paradise during the Camp Fire and worked on the investigation of the Carr Fire line of duty death.

Paladin is his answer: a digital sand table for wildland fire. Enter a set of coordinates and see where the fire will go for the next six hours. Then add hand line, dozer line, or a retardant drop and see what holds.

Topics covered:

Leaving pure research to build tools for the fireline

Lessons from the Camp Fire and the Carr Fire fire tornado

Why fire behavior models feel shoehorned into operations

The Rothermel nomogram and the tradition of putting science in the field

Why 95 to 97 percent of initial attacks succeed, and the 3 to 5 percent that do not

The first-arriving officer's key question: can we catch this?

What a digital sand table is and how it moves from training to operations

Live demo: a six-hour fire spread simulation with tactics on the map

Why Paladin uses physics, not AI, and works offline

Leader's intent, span of control, and export to ICS forms

So take a listen to hear what he has to share.

Head to www.smartfirefighting.com to discover how SFF accelerates innovation for emergency responders, to find out when our next event is, or review our curated resources.

Facebook | Instagram | Twitter | LinkedIn

Chapters

1. 0:00 Cold open: Jackson's background
2. 1:07 Kevin's intro
3. 1:44 Leaving pure academia to build for the front lines
4. 3:17 Why fire research doesn't reach the crews on the line
5. 4:11 Paradise and why this work is personal
6. 5:37 Fires decided in the first hour: making initial attack winnable
7. 7:39 The state of fire behavior models
8. 11:17 The Rothermel nomogram and science in the field
9. 12:44 Perishable skills and repetitions
10. 14:37 95 to 97 percent of initial attacks succeed, and the 3 to 5 percent that don't
11. 15:53 What a first arriving officer is weighing
12. 19:12 What a digital sand table is
13. 21:13 From training to operations, and why trust matters
14. 23:53 The San Jose fire behavior experiments
15. 26:11 Where Paladin fits: command, imaging, and dispatch software
16. 28:54 Live demo: fire spread and tactics on the map
17. 33:23 The tailboard whiteboard and leak testing a plan
18. 36:24 Physics, not AI
19. 37:40 Wind animation, offline use, and bring-your-own weather
20. 40:37 ICS forms, shared plans, and leader's intent
21. 43:26 The fire season outlook
22. 45:37 Wrap up and why not AI for dogs

Questions answered in this episode

What is Paladin and how does it help wildland firefighters?

Jackson Yip describes Paladin as a digital sand table for wildland fire. Before arriving on scene, you can see where the fire will go for the next six hours and test different tactics like hand lines, dozer lines, or retardant drops on the map to get insights into how they will perform.

Hear it at 28:46

Why did Jackson Yip leave academic fire research to build software?

Jackson Yip says the knowledge gathered in academic fire research is inaccessible to people making decisions on the line. He wanted to build something useful for the folks who are hanging it out there, rather than just writing papers for the fire meteorology community.

Hear it at 2:11

How does Paladin differ from traditional fire behavior models?

Fire behavior models answer the question of where the fire is going. Paladin focuses on answering what to do about it, helping decision makers understand if they can catch the fire and where to deploy limited resources most effectively.

Hear it at 8:22

What percentage of wildland fire initial attacks are successful?

Jackson Yip says about 95 to 97 percent of initial attacks are successful. The fires that destroy cities and burn large areas represent only 3 to 5 percent of ignitions, but those incidents follow a power law with horrific consequences.

Hear it at 12:32

Is Paladin based on artificial intelligence or machine learning?

Jackson Yip says Paladin is not AI or machine learning. It is based on applied physics and research, using physically constrained models that allow users to understand the mechanisms of why fire behaves as it does, rather than statistical representations.

Hear it at 36:24

Transcript

Machine transcript of the published audio. Names may be misspelled and some sections are missing.

[0:01] Kevin Sofen: Many wildland firefighters that I know have a version of the same story. You spend days digging a line, you put in the work, and then the fire jumps it, and you fall back to the is the founder of Paladin Industries. He's a meteor-ologist by training, came up through He was in paradise for the Camp Fire, and he worked the line of duty death investigations of the Carr Fire. Somewhere in there, he decided that writing papers about fire behavior was not the same thing as helping people who fight it. What Paladin built. is a digital sand table for wildland fire. You get a 9-1-1 call and a set of coordinates and before you're even on scene, you can see where the fire's going for the next six hours. You can mix and match resources, like use your hand line, or your dozer line, or a retardant drop on the map, and get insights into how it's gonna perform. Welcome to the Smart Firefighting Podcast. Kevin Sofen here, where we cover all things related to the future of smart technology for first responders around the world. Please enjoy the show, and let me know what you wanna learn about next. Welcome to Smart Firefighting, how you doing?

[1:26] Jackson Yip: Good man, thanks so much for the invite.

[1:28] Kevin Sofen: Dude, it's good to have you. And I mean, when I first saw you and came across the scene, I saw you were a meteorologist, you were working in the fire lab and you had worked inside Paradise for the Camp Fire, and oftentimes when I work with people in academia, there's this idea of hey, in academia, you keep getting grants to kind of sustain the lifestyle , and academia used to kind of working alongside academia too, but you were like, I'm not just gonna keep writing papers. I wanna go and actually build software for people who need it on the front lines. How did you have that kind

[2:11] Jackson Yip: I think a lot of it comes from the culture of fire science in general. A lot of the history behind the work that we've conducted in our field is rooted in the forest service rooted in these operational needs and in the modern era around academic research, the funding has pulled us more towards the ideals of the tech industry and things that are much more theoretical in nature as opposed to applied. And so, you know, our lab at San Jose State where I've been a researcher there, you know, basically on and off for a decade, we've always We always tried to make things related to making firefighting safer, making it more effective and understanding the processes that drive wildland and fire. And while I love doing fundamental research along those lines, with the wildfire crisis expanding in the face of historic forestry practices and climate change to some extent, it raises this question of, is firefighting becoming more dangerous? And what is the actual outlet of our research? And so, you know, I have these conversations regularly with firefighters who are, you know, working in wild land areas. I jokingly say, you've read all of our papers, I'm sure. And most of them are like, why the hell? And that whole point is that this knowledge that we've spent so long working to gather is inaccessible to the people who are actually hanging it out there on the line. And so, Paladin is a stab to build into something useful and build it for an audience that isn't the fire meteorology community. It's for the folks who are making the decisions that save the town.

[3:51] Kevin Sofen: And so you've been on the ground, whether it was in Paradise or the Carr Fire, what was something that you saw or something in your brain that was rewired, whether for the worse or for the better in regards to shaping the person you are today?

[4:07] Jackson Yip: Yeah, I think a lot of this comes down to When you see these events happen, there are immediately a number of different things that come to mind as say the California citizen in me, the person who loves the state and the person who loves the lands that we live in. And then the other part of me that resides in sort of the predictive services community of the fire service, in seeing that if we don't enable the firefighters to do their jobs safely, there is no way for us to protect our communities. And so when you look at these sorts of incidents, the line of duty death on the Carr Fire that we're talking about was due to a fire tornado [inaudible 4:46] that ended up registering as an EF3, that's 140 mile an hour winds, basically, on the line picked up a type six F-350 fire truck and threw it a hundred yards and then melted it. And these sorts of things, while they feel extremely random, are to some extent, and I, you know, I had just to say that they are more predictable than we let on, but that data is inaccessible to the people who are out there in the field. And so these same sorts of processes of knowing the arrival times of fires, say from [inaudible 5:20], where the paradise fire started to downtown paradise and making operational decisions based on these sort of actions and how you propose tactics on the board are things that have not been accessible to these first in decision to when we hear about these sorts of, you know, tragedies on these fires, a lot of that comes down to whether it was catchable in the first hour to five hours. And, you know, the question of How do we make initial attack more effective? How do we make them more win Is you're talking about an American city paradise where you're seeing 300-foot flame-lengths, you know, 30-story building sort of behavior rolling through downtown. And I think that we're in this era now where it's hard to conceptualize this if you're in the East, because when have you ever seen a fire that large. But in the West, this is becoming a common reality. We hear about Paradise, we hear about the Palisades, we hear about the Eaton Fire and Quadro. I think it's is unacceptable that we live in an era where we can destroy American cities at such a high frequency. I understand the reasons behind it and I understand the challenges behind turning that tide. But I think there is a point to be made around the fact that when it comes to, say, our work, where we can give better insights into how these phenomena to happen, and then provide that intelligence to the people who are making decisions on the ground. I think it's a disservice to not push that envelope forward to help support these folks at the front lines.

[7:11] Kevin Sofen: And you mentioned the term predictive services for the fire service, and I know within the fire industry there's a lot of these different pre-banning tools or simulation tools, but fire's always been extremely hard thing to replicate, but at the same time we have done it in certain capacities and now technology is getting better with understanding fire science dynamics and then including wind patterns and including all these different variables. But I know there's been some of these in fire services rooted in history in tradition. But I'm curious with the fire service models within the past 10, 20 years that have commonly been used as some of the kind of before and during tools to help kind of guide decisions for initial attack. And you're kind of your knee-jerk reaction on them and the level of them being outdated in terms of some of the challenges they face and kind of the current state [inaudible 8:10]

[8:22] Jackson Yip: as a concept. They've been research tools. They were architected for the purpose of understanding the dynamics of how fire propag talking to incident commanders and you talk to you know first in decision makers and decision makers at complex incidents as well, they really are less concerned with the fire dynamics so much as they're concerned with answering the question of can I catch this, where fire behavior models answer the question of, where is this thing going to go? Plan building and the sorts of things that we're trying to push answer the question of, what do I do about it? And there is a fundamental difference between those two things in not only the design but also in the technology that sits behind it to enable non-analyist users and people who are having to make knee-jerk reactions in the field based on their recognition priming to be able to save the town? Do But interestingly, in the fire service, the community is aging. We're losing these brilliant tacticians that have existed throughout the end of the last decade, and they're retiring now, and there are people who are going in and making decisions on these expanding incidents, who are young and don't have the repetitions that some of these other folks did. sort of to, lean into the other side of it, you know, the design process behind this, it's really critical that we ask that question too, is how do we make it usable for these people? The community has become much more tech native, folks are my age, you know, I'm 31 now, and you know, all of us have a phone in our hands, all of us have a tablet in the truck and the idea of baking these technologies into a single pane of glass, so that you see the entire situation in front of you, so that you can react because that's kind of how academia is. And some of these tools that have historically been used are brilliant for their time because they And the one that I'm mostly pointing to is what's known as the nomogram. And you hear about this in your S-490 courses. And it was developed by Dick Rothermel from the Forest Service Fire Lab, I think he was in Georgia. And the thing that it provided was, instead of me having to read four papers about how fire behavior works, I have a single chart where I can draw all the conditions that I'm seeing on the ground, and see what the rates of spread are, see what the flame lights are, and understand the dynamics. And obviously, this was pre-iPad. This system was developed in the 70s, but it takes the same philosophy of taking the science and putting it in the hands of the people who are making decisions, so they're not left in the cold.

[12:04] Kevin Sofen: So right now, I'm curious, if you could walk me through what actually happens. 911 calls in, there's smoke, and you're their first-hand officer. What are you doing and how are you now sort of making these decisions in regards to where to position your people, where to position your resources, what to what to do next? I mean, beyond just kind of, relying on your gut and kind of, you know, putting your finger up and checking out the winds. I'm curious, what's being done now?

[12:32] Jackson Yip: I don't want to discount the expertise that's present in the fire service. There are some of the most brilliant tacticians you could ever think of working They'll be looking at the smoke plume to see if it's standing or if it's blowing over, which infers basically what the convective behavior. So how And then they're looking at what the assets that are the infrastructure, the various types of landscape features are present where this fire is going and we're going to try to defend those things. And basically, the question is this going to blow up into something more or am I going to be able to catch it? And when a first in commander is walking in there, and I think it's probably more useful to talk directly to one of those individuals because like I said, I've never been that person. and I'm a meteorologist, but I try of those people.. They're extremely task saturated, they're extremely stressed from this expanding incident, and they have limited resources to deploy on the map as fast as they can. And so when you look at the initial attack failures, and these aren't things that are necessarily caused by a misstep or something like that, you see that basically resources couldn't get there in time, they didn't go chose the wrong flank based on what the fire behavior was, or conditions were so extreme that there wasn' t any way for them to catch them in the first place. And so the thing that I think is, you know, lost in the general public is how good we are at initial attack in general. About 95 to 97% of initial attacks are successful, and the fires that we hear about that we hear about destroying cities and you know burning down half the Sierra those sorts of things only represent three to five percent of the ignitions that occur. And so the thing is that when you get down to that other three to five percent of ignitions it's what we call in science and math of a power law is you have this long tale in of horrific incidents the ones that nobody ever saw coming and within there aren't. We can basically say that there are incidents that we can pull back from the brink and incidents that weren't savable in the first place. And having a tool that you can pull those incidents back from the brink means you can trim off more initial attacks that are winnable and help these people who are already making great decisions

[15:35] Kevin Sofen: When thinking about how, again, these different first-arriving officers and like the different decisions that they need to make, and like they're very well trained, and there's all these different factors that are happening and your heart rate goes up and you're trying to do what's best for you and your team and obviously the community and trying to reduce the fire spread, I'm curious, what are the different things going on in a first arriving officer's head that they're thinking the time window so short, the hope to be able to make the best decision is obviously critically important.

[16:34] Jackson Yip: Yeah, I mean, the first thing is safety. If a team can't engage a fire because the fire behavior is too extreme, they're not going to do so. And when it comes down to that philosophy, this comes from what are known as the 10 and 18, the 10 watch outs and firefighting orders to basically inform folks on the ground what the best decision-making process is. And so when you get down to safety protocols around this, it's all driven by current and expected fire behavior, and fire behavior is driven by the fire triangle, namely terrain, weather, and fuels. So understanding those dynamics are some of the most critical parts of this. And as you said before, when you're looking at a wildfire, it's one of the most dynamic environments possible. So, how can we expect someone who's rolling in trying to stop this horrific problem that's going to be running calculations to try to understand where the fire is going to go relative to their lines when it's super dynamic. So, you know, when it comes down to what's in their mind, the question is can we catch this thing. That's the biggest question at hand beyond safety. And a lot of the tools, the tools out there don't help provide the opportunities that exist on the landscape. And so, you know, Paladin in its inception was intended to for one help point out weaknesses because firefighter safety is paramount. But the second part of it is that we help point out opportunities. So did you know that there is a road there? Did you know how effective that ridge line would be? There's a side of it that, early on when we were talking about what we wanted this to be, we wanted to figure out a way that we could make the strike of a Pulaski be twice as effective every place where it's put. And this comes down to the mindset of, once again, this is not a very controversial opinion these days because of how much money has been put into this, is that you could buy all the aircraft in the world, and still put them in the wrong place. And it comes down to the expertise of that first indecision maker to be calling in those resources to put them in the most effective place possible

[19:16] Kevin Sofen: little sand table. I'm curious if you could sort of communicate what that means in your words [inaudible 19:27]

[19:46] Jackson Yip: ships in it, to build scenarios, lay out the overall decision game, and then leave incident commanders to practice what they would do in the instance of a fire-specific terrain. You know, it doesn't take a researcher to tell you that a box full of sand is hard to bring to an incident. And this idea of how do I understand what decisions to make and how do I leak test my plans comes from this idea of taking a concept that's already understood in fire service and modernizing it into something that's useful in the field. So, you know, when we're talking about what Paladin is, it's not only a training tool, you can use it as a training tool because you can, you know, put an ignition on the map in a place where you've had a fire in the past, see where it's gonna go, we're gonna share this between you and me, you're gonna be a division supervisor, I'm gonna be an incident commander, and we'll have another friend who's another division supervisor, and they'll put their tactics on the board, we'll see how it adapts, and then we can have a conversation around if we made the best tactical decision we could. The intention is not to tell firefighters what the right answer is. The intention is to help show what will and will not work based on current and expected fire behavior. Falling in line with the doctrine is one of the most important things. And when you look at how this becomes useful, we're one of the first stabs at bringing a tool to the community that can walk across from training to operations. And so when you're talking about when you get the call out and you have the latitude and longitude, now this tactical decision game that you've been practicing on is the thing that you see in front of you with real conditions, with real circumstances, and real firefighting to make these decisions on. And so, you know, we, you know, I posted about this week is talking about trust and talking about why, you know, trust matters so much. And we were conducting these fire behavior experiments in San Jose to really understand what the accuracy and shortfalls of our system is. And very hopeful to say that this has been a very validating experience and the experience we've conducted and the data that we've processed thus far. But when you know, when it comes to building tools that are useful in the like in the fire ground, in the time compressed environment of initial attack, it has to be familiar, it has to be fast, and it has to fit into the mental model of how firefighters already think. And so we went for the digital sand table mindset, because you can basically just pull up the iPad, put an ignition on the map where you see it, or you have an auto-ignition present on the and then you start adding your tactics to it and see where the fire spills over, or where you can catch it with minimal effort. And this idea is that we're trying to represent this ground truth as closely as possible by and after action and understand what's happening

[23:39] Kevin Sofen: radio and all that, but I'm curious kind of how you see Paladin and how it fits in and even kind of what was from this exercise at San Jose this past week, which had a lot of different players and partners there tell me more about that.

[23:53] Jackson Yip: So I guess kind of to start on the San Jose experiment, I personally think that San Jose is the center of wildfire research these days, namely because of the wildfire interdisciplinary research center at San Jose State. And when it comes to owning that, it comes down to the fact that acre burn experiments to understand fire dynamics and in multiple different ecosystems in its brain. The intention was how do we align the airborne systems that the best groups in the world are using to quantify fire behavior from the air when real fires break out, to So how do fires work, to understanding how we model it? Basically the question of how do we represent how fires work in computer simulations. And so we put together this experiment where we had our friends at Santa Clara County who were working on a thinning experiment, bring down about 25 cubic meters of Santa Cruz Mountains fuels where the CZU complex fires happened in 2020. We cured them a number of days and then we burnt them in a highly instrumented environment with multiple different camera systems viewing them at the same time with targets. So what that provided us was a place where we could observe fire behavior, observe camera calibration, and then use all the observationalational data to run analyses on predictive models to really understand where they fall apart. And so, you know, the findings that are coming from these experiments help us align the fire ground intel community around common truth. and truth is the thing that we are all in the market trying to provide for the person who is having the most stressful, dangerous day you can imagine because that's the thing that they ask for from our community. And so if we're not providing that, then we're failing them. When it comes to the second part of your question there, you'll have to remind me what the second part was. When it comes to the second part of your question there just kind of how Paladin's fitting into the greater ecosystem and yeah so like so like I tend to think of it in terms of what questions each group asks and so like we were talking about with fire behavior models you asked the question of where is this thing gonna go when it comes to operational suppression tools like your fire apparatus and helicopters and those sorts of things it's the you know how do I how do I stop it or more on the lines of you know what can I put on the ground that side of it we sort of fit into it along the mixture of you have commanders who are making decisions you have resources that are your chess pieces and there's no one really sitting in between those places basically say okay how do most effectively use the resources on hand And so a lot of the focus that we've had is not only from the how do we make a better tool that's the predictive sandbox sort of mindset, but also from the question of how interoperable is our system with other systems. And so we pair up with command software, we pair up with imaging software, we [inaudible 27:25] [inaudible 27:26] And so being able to provide additional guardrails to those individuals is something that we really focus on.

[27:57] Kevin Sofen: So we've been talking about it for a little bit. Is there any ability we've ability to maybe see a little screen share or see it in action? So it's great to see it and talk about it and I know you've had some of these different users actually experience it and use and use it, but to actually see it moving and I think hearing you talk about it, where again, fire service members are some of the most well-trained, well, best prepared individuals, but still if we have technology that can help them make better decisions and have better testing on where they're going to deploy their limited resources, why would we not use that?

[28:36] Jackson Yip: Of course.

[28:38] Kevin Sofen: I also guess in the medium term, the name Paladin itself, where did that come from?

[28:46] Jackson Yip: It's sort of a funny story on that one, because it's my middle name. It's literally my middle name, but the connotations behind it obviously are different, And one of the things that really mattered to me when starting this company was this idea of helping heroes be better heroes and focusing on this idea of not being the person who stands in front of this thing necessarily, but being able to be the support role that our community really has strived to be for the folks who are making these decisions out in the field. So this is Paladin. And I'm just zoomed into basically Southern California here. I can see Pasadena off to the left here, San Bernardino. And so if we go into a zone like back in here, for example, the thing I can do basically is if I have coordinates from dispatch or I drop them on the map because I just got dropped in there, I can put this on the map and say, okay, well, here's the terrain that I'm operating in. How is this situation going to evolve? And so from that minimal data that I'm providing, Now, of course, this is one that's right in fuels that are not going to do anything. a lightning-strike or something like that. I can place this on the map and say okay where is this going to run to and this is a pretty low-intensity one we find one that's gonna rip. I mean it is real-time in terms of what it would [inaudible 30:12] and all of the watch out zones that you'll get from this. So what you're looking at here, I'll turn off the opacity so you can see the perimeters themselves. You have basically terrain corrected fire behavior that's showing you all of the watch out zones across this slope as a result of the current fuels, current weather and current terrain. So if I turn this back up again, the thing that you'll see is, okay, well, even just upon first glance, Where are the places where I should or shouldn't put my firefighters? I can see all of the terrain runs around my firefighters. I can see all of the terrain runs And so this is what firefighters would get before they even arrive on scene to say, Okay, well I can see where my extreme fire behavior is going to be and I can see where my opportunities are going to be. Something you might see right here is I've reduced fire behavior through this area here. And so the question becomes can I catch this along this ridge line and maybe the question is okay. Well, I'm going to fly. an, you know, an aircraft down through here. And now I can add that to my plan. And so I have these opportunities to basically say, okay, here are my watch outs. And now I can say, okay, well, I'm gonna, you know, this is a very optimistic, retardant line from a very good pilot clearly, that I can then ask, okay, well, what happens if I put those tactics on the map? . And so here it's spreading and the question becomes, Okay. Well, you can see on this side here that it bled through. So I'm going to need more for this purpose. So I'm just going to go to town here. This is not all realistic in terms of firefighting tactics, but I'm going to put in a hand crew line. I'm going to put in a very optimistic dozer line across the side of this mountain and then put in a hose lay across the top of this. and we can see what will and will not hold basically based on these tactics.

[32:32] Kevin Sofen: Yeah, based on all the different tools in the toolbox that have incident commander working with this different limited resources here you can kind of look at where based on the ridge lines and based on current fuel where you have the best chance to actually fight this thing as opposed to sort of be saying we should drop the retardant here based on this is maybe where the fire is hottest or where the fires rip and it's more so of course knowing working with the fire working with the land and finding how can we contain this fire in the best way possible by looking at what's where the where these you're kind of here just tinkering and testing with different variations they're almost like you're kind of playing a simulated game of chess and just kind of playing it a few times and tinkering with what's the best plan going to be

[33:17] Jackson Yip: yeah exactly and so it comes down to this idea of there on basically every battalion chief truck you see in, you know, in California, there's a white board on the tail board. And this is where resources are reconciled. This is where plans are made. And so, you can build a plan exactly the same way as you always have in a sand table like this. But the question becomes, okay, well, is this plan going to work? You know, you can be as optimistic as you please, more and more resources here and I try to like make this more and more effective, answer the question from a mechanistic level of is this the sufficient response given the current situation.

[34:06] Kevin Sofen: Yeah and is putting this extra hand line here or getting this extra airdrop here going to do anything to help us accomplish our goal.

[34:16] Jackson Yip: Yeah and this is one of the things that I fundamentally baked into the philosophy of the company is. The tacticians who are using this know what to do, but they don't know if it's going to work until the fire burns up to the line that they put on the map. And so sort of to your point in the intro of, you know, one of our software developers is a former hot shot. You know, he tells these stories of, you know, he was working the Monroe Canyon fire last summer in Utah of, they had spent 14 hours digging line winds would whip up in the afternoon, jump the line, and they did that for a week straight. And it's like, you know, it comes down to when you're making decisions, it's a tool that can help guess and check what your decisions are doing. So you can basically now game out this idea of, okay, well what is it going to take to stop this or do I need to just fall back? because oftentimes the idea of falling back to the next ridge line was the thing that had to happen anyways so now we don't have to be chasing our tails around in this spot we can say okay well this is what the current and expected fire behavior is based on you know the current fuels current winds current terrain and we can make strategic decisions instead of basically chasing something down that's not necessarily going to succeed in

[35:42] Kevin Sofen: As you were talking through and I saw those simulations happen, it made me think, and without you necessarily releasing the special sauce of Paladin, but I'm curious how that is working. I mean, you have the map, but then there's different fuels in there, and rains happen, and there's historical conditions, there's stuff that happened in the last week, and then there's stuff that's happening in real time, and then you shouldn't be, dozer line and hand line, those both have different types of level of effectiveness, and so you're kind of dropping if this type scenario is based on what is reality out there, how's it work?

[36:24] Jackson Yip: I can't necessarily go into the mechanics specifically, but I think that the important thing to say is that it's not AI. It's not machine learning. It's a tool that's based on the research that we conduct as a state and we ourselves are applied physicists. When it comes down to the mechanics of how this works, we understand the mechanics well enough to build a tool that's representative of the real world instead of being something that's statistical. So when you look at AI fire models, you ask the question of, why did this happen? you get a shrug. There's no answer as to why these things happen. When it comes to a physically-constrained model, you can actually tease out the mechanisms [inaudible 37:10] And it represents the mechanistic behaviors of the system rather than the statistical representations of that system.

[37:38] Kevin Sofen: That's great. And then as I'm also pulling into full screen myself, I see up top here, behind this test text and you've kind of gone through. There's wind animation. Talk me through some of these other nuances of buttons. And again, being able to replay the simulation or tactics versus current WX, custom WX, just click through everything else here that someone can potentially experience while using the software.

[38:07] Jackson Yip: Yeah, totally. I'll just walk through some of these. There are some settings that are very specific to our users, and for more details on those, we would defer to a demo. But one of the things that we wanted to be useful, [inaudible 38:21] something that reflects fire ground conditions in a way that is realistic. And it needs to be something that is interpretable by an S290 audience or up. So the idea with this is that it fits into the pocket of somebody who is a Type 4 Incident Commander or higher. I guess a Type 5 Incident Commander is more than capable of using this as well. [inaudible 39:17] that are out there, but it serves the specific purpose of how do I make the best decisions. That's the idea behind this. And so you can lay out basically all of the different components. that you care about when it comes to an incident. - so if I have a specific objective that I want to shoot for, I can just say, hey, this is what we're trying for here. I can lay out my divisions. So if I want to say my division breaks and label those, I can basically start making-a plan accordingly. And so to that effect, we fit directly into the ICS pipeline, And when it comes to building a plan, you can just export it to common ICS forms, which creates an offline shareable. So that's just exporting a nice PDF and giving a plan of, hey, here's the fire, here's what's going on. And based-on our best- [inaudible 40:57] is the idea of, you know, if I am trying to accomplish an objective as the CEO of Paladin, everyone in the organization below me should know what that objective is and know how they fit into my intent of solving that problem. When it comes to span of control, you know, there is there's a certain aspect of an incident becomes too large for the incident commander to know the names of everybody, so to speak. That's the easiest way of putting it. But Span of Control basically says that an individual cannot manage more than five people on their duty as an individual. So the reason that ICS is laid out the way it is, and the reason that command structure is laid out the way it is so that you don't exceed the Span of Control a such that you lose control of your management cycle. And so the idea of having shared cognition means that you're communicating both leader's intent in a concise way as well as making sure that everyone is on the same page as far as span of control the entire time. This comes down to, you know, aligning assignments and making sure that people are where they need to be as well as reconciling assignments so that there aren't people on the line who weren't assigned and there aren't people who are missing that you thought were one place and now they're somewhere else. This idea of making sure that everybody is known where they are and making sure that they're accomplishing their goals is one of the critical parts of building a system like this.

[42:51] Kevin Sofen: And it gives a lot of comfort of knowing what you should be doing, especially and especially whether you're managing someone or being managed, it's good to know exactly what to do. It also gives you comfort knowing that this is going to going to help create the most amount of impact and best outcome. Granted, this is still simulations and a tool and things can change, but this is something that's based in science and research, and it's the best the best available tool to help better make decisions. So I'm excited about what this can do and how it's helping and I'm curious now as it's about to say August, but September 20-26, what are you excited about from both kind of a technology perspective and kind of an upcoming Paladin perspective over the next six to 12 months. What's on your radar? What are you looking forward to? What are you excited about?

[43:46] Jackson Yip: Yeah, I mean we're sort of standing at the door to the wind-driven fire season in California right now. So usually we get these events, Southern California, we call them Santa Ana's in the northern parts of the state. We call them Diablo winds in between their sundowners and all these other types. We're looking down the barrel of a pretty difficult season, the sorts of fires that we've seen across the west this summer in the Pacific-Northwest as well as in Colorado that unfortunately have claimed a number of firefighter lives in the line of duty. We're looking at similarly scary conditions here. And I realize that you asked about what I'm excited about, but the thing that I'm excited about is contributing as someone who lives and loves California to help contribute to the decision-making process for these to hopefully pull more of them back [inaudible 44:39] that you wonder if folks are going to adopt a new technology that they've never seen before. And then they very clearly start seeing all the places where it plugs into their day-to-day life. And, you know, when it comes to Paladin, we had that experience in the spring. We are, you know, deploying with a number of agencies right now. And when it comes to someone making decisions makes their day better, and that's what I'm really looking forward to this season, is the idea of being able to contribute.

[45:37] Kevin Sofen: Right on Jackson, well I appreciate your journey from academia and meteorology to building software that's purpose-built and actually helping both first responders and the initial attack process just be better, faster, safer. So thank you for all your. And I'm excited to continue to follow to be part of the journey to see where Paladin goes. And yeah, my kind of final thing I ask people on the pod and sort of a question I left field, but you being somewhat of a philosopher and stoic yourself, I'm sure you've got something up your sleeve, but is there any favorite quote or words of wisdom that Jackson lives by as you, the individual, and or anything that's sort of the Paladin, words of wisdom, company motto, or anything?

[46:24] Jackson Yip: You're catching me off, actually. I think that it's important to sort of live what you believe, and I think that in a world where, we live in Silicon Valley, so we're surrounded by every AI for dogs company could ever imagine. And there's a certain amount of I don't want to build things that aren't useful to the community that I care about. And so the idea of walking the walk is entire lab is trained firefighters to be on real incidents. The idea of everything that we build needs to contribute a use to the operational community. And if we're not, you know, playing those games and living and living, honestly, with that mindset, then we're failing. And so, you know, we're in the market of trying to make it safer and more effective for firefighters. And the things that fall out of that for our community as Californians, and say like the business side of that, are all things that are only enabled by us making it safer for firefighters to do their job.

[47:46] Kevin Sofen: Well I'm sure happy that you are putting your energy and talents towards that and not AI for dogs because well, I do love my dog. I think there's enough AI this AI that for dog services that is not as important as what you're working on. So well said, and thank you for the tireless work that you're doing for the fire service.

[48:08] Jackson Yip: Yeah, of course, wouldn't have any other way.

[48:11] Kevin Sofen: Well Jackson, look forward to talking again soon across and past and keep up the great work at Paladin.

[48:16] Jackson Yip: Thanks man, good to see ya.

[48:18] Kevin Sofen: Take care, see ya. Thank you so much for listening to the Smart Firefighting podcast today. If you enjoyed what you heard and got any value, please drop us a rating, leave us a comment, or reach out to us on social media. Have a great day and together we can advance the future of Smart Firefighting.