In this episode, Drew talks to Victoria Hutchinson, of the NFPA Research Foundation, and data guru from Netage BV, Bart van Leeuwen. Both Victoria and Bart are industry veterans. Victoria grew up in a fire protection family and is a design engineer as well as being the project manager for the NFPA Research Foundation. The NFPA Research Foundation is a different not for profit organization spun from the NFPA with the overall mission to manage and communicate research that informs the NFPA standards, and are going to inform the codes. Bart works for Netage B.V. and is a part time fire captain in Amsterdam, Holland. Netage B.V. uses smart data to help protect first responders by supplying them incident data and other data including that created via AI. The trio discuss the ITM Project in detail from what it is to why it is important, to how it works, to all the various parts and pieces. There is a lot to unpack! There is really no consistent way that all the players in fire protection create and store their data. The ITM Project has been working with 40,000 fire inspections and trying to find a data model to combine the various data streams included in those 40,000 reports. The goal is to make that data findable, accessible, and available to all relevant community members, from the contractors who perform the inspections, to the building owners, to insurers and Joint Commissions/AHJs, and to companies like Inspect Point, who provide the means to collect, share, and store fire inspection data. NFPA Research Foundation is on the NFPA website at https://www.nfpa.org/News-and-Research/Resources/Fire-Protection-Research-Foundation There, you will also find the link to submit a request to be included in the research process. Netage B.V. website is https://netage.nl You can reach out to Victoria via LinkedIn or via email at email@example.com You can reach out to Bart via LinkedIn, his company website (Netage B.V. above), via his Blog blog.netage.nl, or via his Twitter account, @semanticfire
Meet Victoria Hutchinson (2:46)
Meet Bart van Leeuwen (4:20)
Smart Fire Fighting (6:04)
Software Platforms & Solutions for AHJs in US (7:28)
What is the NFPA Research Foundation? (8:26)
Four People Covering 50 Topics at Once (9:30)
Bringing Research Into Practice (10:10)
Projects the NFPA Research Foundation is Working On (11:36)
Anyone Can Get Involved — like YOU! (12:29)
What does ITM Stand For? (15:51)
Why is This Needed? (16:08)
Did This All Stem From Fire Pumps? (17:48)
Everyone Wants the Analytics (20:25)
Fire Pump Churn Testing (22:48)
Why Are NFPA 25 Meetings Funny? (23:42)
Data Grabbing Is Tough (27:46)
Who Owns the Data? (30:49)
Sharing the Data (35:09)
Performance Based ITM (36:46)
Collected Data from 40,000 Fire Inspections (40:32)
Will the Results of that Data Analysis be Game Changing? (43:05)
Report of Project (43:10)
Next Steps (43:43)
Can We Create a Uniform Data Model From Diverse Streams of Data? (44:26)
Will the NFPA25 Update in May ’21 Use This Data? (46:34)
Get Involved! (47:40)
New Direction Coming (48:41)
The Data Drinking Game! (48:48)
Quick Response Round! (49:30)
Nerd Question Alert (50:39)
Shout Out to Jeff Norton!(51:25)
NFPA Bedtime Stories for Victoria (52:40)
Novec 1230 Systems (54:05)
There Will Never Be A Fire At A “Normal” Person’s House (55:10)
Fire Captain Bart Has Great Stories Not Suitable for Podcasts! (55:35)
Where Can People Find You? (55:40)
Wrap Ups (57:28)
Discussed in this Episode: NFPA Research Foundation, NFPA Codes and Standards, Inspection Data Collection, Fire Fighting, Emerging Technology in Fire Protection, Inspection, Testing, and Maintenance, Fire Pump Performance, Analytics, Fair Data Practices
Drew Slocum: (00:10):
This is episode 29 of the Fire Protection Podcast, powered by Inspect Point. Today, my guests are Victoria Hutchinson and Bart Bart van Leeuwen. They’re both with the N F P A Research Foundation. Uh, Victoria works actually for N F P A. Uh, BART is actually a consultant working on the I T M inspection testing and ma uh, data, uh, project, uh, for this N F P A Research Foundation. Very intriguing conversation. Um, and what N F P A kind of works on with the N F P A Research Foundation. Um, it’s kind of a, it’s, it’s, it’s a bit different than just the regular nonprofit N F P A. There’s a, uh, I, I learned a lot of what the N F P A Research Foundation is doing, and then we get into the actual whole I TM and inspection, testing and maintenance, um, project specifically, they’re working on and trying to come up with a better way to gather data for fire protection systems out there, mainly in N F P A 25 and N F P A 72 for, for fire alarm and, uh, sprinklers.
But both have, uh, rich backgrounds with fire protection and data. Uh, really, really cool to talk about that with them. So, again, hope you enjoy, please just subscribe to the podcast. Uh, one little tidbit of information I wanna, uh, share with you before we get started is, uh, the inspect point. Uh, fire Protection Podcast is gonna make a debut live at the upcoming, uh, Nafe events in, uh, Las Vegas and Atlantic City. So, uh, inspect point’s gonna be there at the trade show, at the conference, uh, being there like we normally do. But I’ll also be conducting a live fire protection podcast from the expo in the conference floor at both Las Vegas and Atlantic City events. So stay tuned for more. And, uh, again, please subscribe and share. Thanks. So, uh, thanks Bart. Thanks Victoria for, for joining me on the Fire Protection Podcast today. And, uh, I’ve been meaning to have you guys both on here for a while, um, been involved with some of the data projects that, uh, that N F P A has been working on, and, uh, really excited to kind of hear a little bit more with that today. So, uh, welcome.
Victoria Hutchinson: (02:36):
Thanks for having us.
Drew Slocum: (02:38):
Thank you. So, um, yeah, we, we’ll get into the, the, the TM and the data project that NFPAs working on. But, uh, I guess let’s just start with little intros and backgrounds of, of, of where you both come from. Uh, Victoria, do you wanna you wanna kick it off?
Victoria Hutchinson: (02:58):
Sure. Um, my name is Victoria Hutchison. I’m currently with the Fire Protection Research Foundation, which is the research affiliate of N F P A. Um, I’m a research project manager there. I’m a little bit on my background. Um, I have a bachelor’s and a Master’s in Fire Protection, engineering and Safety Technology. Um, I come with past experience, um, in the fire protection engineering world doing, um, design and particularly inspection, testing and maintenance as, as well, um, as some of that background before coming to N F P A. And then here at, um, the Research Foundation. I manage a lot of the projects, you know, related to NF P 13, um, you know, active, uh, active safety systems. Um, I’ve started doing a lot in the data and analytics space as well, um, in some of the merging trends and, and technology. So, um, what we’re gonna talk about today is, is one of those, those projects that, that I’m in charge of and kind of brings in my, my background a little bit with, um, the ITM experience as well. So, definitely interesting stuff and, and excited to be here. Uh,
Drew Slocum: (04:09):
No, no, thanks. Thanks for, uh, yeah, you have a wide, wide background and, uh, uh, again perfect, perfect for what you’re doing now. So, uh, Bart, what about you?
Bart van Leeuwen: (04:22):
Yeah, thank you. Uh, my name is, uh, bar Van Lavin, uh, from Net Edge. Uh, we’re Dutch company. Uh, and I was introduced myself as wearing multiple hats and helmets. Um, I’m here in the capacity of net Edge. We’re a small company specialized in smart data for smarter firefighters, um, and, and have been within the industry for 25 years and focusing on the fire surface for the last 15 years, almost, uh, and trying to be on the edge of, um, academia and engineering. So take, take some of the research on, on, on data management and, and knowledge graph and graph structures and, and AI and all that. And, and try to bring in engineering touch to that, and then bring that to, to the, to the fire field. Um, I’m also a, uh, part-time captain of the Amsterdam Fire Department. So I have, um, more than 25 years of firefighting experience in the meantime with 20 years at the Amsterdam Fire Department.
Um, so that brings for me a lot of practical, uh, knowledge from the field. Uh, and I hold a guest research position at the Free University of Amsterdam, uh, on, uh, the data, um, uh, the, the knowledge and media group where we look at human-centric data, human-centric ai, and, and explainable ai and all, all these kinds of things. Um, I got into this project through some other work we’ve done, uh, with the N F P A, where we used the graph model, um, as a, as a means of trying to converge various different data streams into one single data stream, uh, to be able to do some analytics, uh, on that. Um, and that is basically what this project, um, is about.
Drew Slocum: (06:04):
Gotcha, gotcha. Um, what, I guess, smart firefighting, what, uh, what part of smart firefighting? I mean, there’s, uh, it seems more, more on the, the equipment they’re wearing or just the, the data When, when the incidents occurred coming into a, coming into a building.
Bart van Leeuwen: (06:25):
Well, well, being a, a sort of an urban firefighter myself, uh, originally we started looking, okay, how can we make sure that the, um, sort of responding crews have the proper information? Um, and, and specifically looking at that from a situational awareness perspective. While we were going into this project and started doing this project, they started looking, oh, but, but if you have, if you look at some of the, the regulations with fire protection equipment, they actually play a role in the way we fight fires as well. Sure. Uh, and if you go from fire protection equipment and smart buildings, um, then how could we better interact with fire protection equipment? You know, what, what’s the, what’s the future of all these technology interconnected, and then being unreal to a fire, actually being able to see what’s going on in the building, et cetera, et cetera. So that’s where the link for me is. But I have to admit, the smart firefighting for me, uh, has initially been specifically the, the, the response operations.
Drew Slocum: (07:26):
Oh, interesting. Yeah, I’ve, I’ve, uh, I’ve chatted with some recent companies here in the States, uh, and, and doing a lot of research with it, but there’s a lot of, you just have different software platforms and solutions out there for authorities having jurisdiction, you know, hjs in the US for emergency re, you know, management response. And, um, I’m, I’m having first do, they’re coming on the podcast, Andreas Hoer, he’s coming on the podcast to, to discuss their solution. But, um, yeah, very topical these days. I’ve got a, they’re coming on here soon. So lot, lot of fun stuff happening in, uh, the data world of, uh, fire firefighting and fire protection, so,
Bart van Leeuwen: (08:10):
Drew Slocum: (08:11):
So, uh, again, chime in whenever, uh, either of you, but, uh, so you have N F P A National Fire Protection Association, and then there’s, explain to me what the N F P A Research Foundation is.
Victoria Hutchinson: (08:30):
So the Research Foundation is, is a separate nonprofit organization from N F P A. Um, but so our overall, um, mission and vision is to essentially manage and communicate research that informs the N F P A standards and development process, um, but also is in support of the NFPA mission. So it was formed back in 1982, um, as NFPAs Research affiliate. Um, and that was really, it was started, um, based on the fact that they, the codes and standards process identified a gap in, in data, really. Um, they, they said we’re, we’re trying to push safety forward through, through the codes and standards process, but we really need research and we need more data to start informing, um, some of these decisions that, that are going into the code. So the foundation was, was started, um, at that time. And really, we, we primarily facilitate research.
So we have Bart on the call with us today, and that’s kind of exemplifies kind of our, our model. We’re, we’re a pretty small group. Uh, there’s four of us at the foundation that, that manage research projects. Um, but at any one time, we almost always have about 50 different projects going on at one time. Um, so in those, you know, cover up cross section of, of topic areas, a lot of them support the codes and standards in some way. Um, some of them are other, you know, emerging issues just in the fire protection community, um, data, data collection, analytics. That’s become a, a priority, um, topic area as well. So very vast topics, um, that we address. But the other thing that the foundation does, you know, for our projects is we really try to be that, that link of bringing research into practice, um, it, all of the research is, is very applied in the idea is that it can be, um, you know, directly implemented by stakeholders to help them manage their risks to, you know, inform the codes and standards process, inform a performance based approach to something.
Um, and that additional, like validation level, the foundation for all of our projects, we have, um, a project technical panel of subject matter experts, um, that review and, um, you know, help define the, the scope and the direction of the project, uh, the technical rigor of it, um, and evaluate that before it goes out. And just for the projects that we have ongoing, right now, we’re working with like 750, uh, external people, um, ACR across the world, um, to help with these projects. So, you know, we try to be that connection to the fire service. We try to, you know, bring in, you know, cross section of experience to help inform these projects. So, you know, just like on this i t m study, you know, we have people with a lot of experience in, in TM and fire protection systems and the codes and standards, but then it’s also getting that data perspective as well. So a lot that’s a lot that’s going on, um, a lot that that goes out. But the ultimate goal of the foundation is, is to continue to produce research that, that helps move the industry forward.
Drew Slocum: (11:42):
Now, it’s, I I, I dove into it a little bit and I got it up right now, but all the projects, you know, you’re working on from obviously the TM one we we’ll discuss in little bit, but just the, the firefighting foam from the firefighters mm-hmm. Point of view to the, the aircraft hanger protection. I know there’s been a big topic. I’ve got an episode coming up soon with that, and I, female firefighter ppe. That’s, that’s very interesting too. And there’s some really cool topics in here. Um, how do you, I, can you get involved with these if, like any, can anybody get involved or is it mainly just kind of the N F N F P A research team picking out different, um, uh, experts in the industry?
Victoria Hutchinson: (12:29):
No, any, anyone can get involved and anyone can submit an idea. So we actually have a, uh, project idea submission portal on our, um, main page of our website, nfpa.org/foundation. And that’s a portal to where anyone in the world that has a research need, um, that needs to be addressed, can submit that to us. And it’s open year round. We, we collect projects year round. We, we go out for sponsorship. Sometimes we find partners and, you know, try to do larger studies to, by getting federal grant funding, the, the mechanism in which we implement it, um, is, is different. But the research that we do is primarily all informed by, by the industry. Um, so it’s rarely something that, you know, just, just, we come up with. It’s, it’s what we hear as needs from, from the industry, talking to people at conferences or people reaching out. We have research planning groups, um, you know, we have a sprinkler alarm, uh, electrical research planning groups. We also have property Insurance Research group, which is a paid research consortium who contribute a certain amount of, of funding every year to do projects that, um, are of interest in the insurance industry. I just launched another one for Energy Storage Research Consortium.
Drew Slocum: (13:47):
Saw it. Yeah,
Victoria Hutchinson: (13:49):
Emerging technologies. So, you know, a a lot of things going on a lot of different ways. We also send out a request of the technical committees every year, um, and F P A gives, you know, a small amount of money for us to conduct research to help, um, fill the needs of the codes and standards process as well. So that’s evaluated every year. So there’s a lot of different ways that, that we get the ideas, but it’s, it’s definitely all all informed by, by industry. And then, you know, just through our, our partnerships, we work with a lot of universities on, on different grants, like you mentioned, the, um, the female firefighter, P p e. Yeah. It’s actually a really interesting issue, um, that hasn’t got a lot of light, um, that’s actually posing some risks to two female female firefighters. And that’s one where Florida State is the lead, but the foundation is, um, supporting and trying to bring that connection, um, from the fire service to that, to that research team. So a lot of different ways that people can interact with the foundation. Um, um, a lot of ways, you know, you can sponsor a project, we can, um, if you have a certain expertise, you can reach out to us and we can consider you for, uh, technical panels for different projects. Um, or we can, you know, partner on research or you can, you know, be like, BART and apply, you know, respond to an R F P and, uh, do the research, um, for, for different projects. So, yeah.
Drew Slocum: (15:14):
That’s cool. Lots
Victoria Hutchinson: (15:15):
Of opportunities there.
Drew Slocum: (15:17):
So, Bart, is, are you only involved in the, the TM data exchange, or are you involved in any others as well?
Bart van Leeuwen: (15:24):
No, currently I’m only involved in the TM data exchange. This is the first time I worked together with, uh, with the research foundation. Um, but I, I think, um, uh, the corporation was really nice, so I’m looking forward to do more, uh, especially when, when it’s data related, because that is our niche, obviously.
Drew Slocum: (15:41):
Sure, sure. I mean, there’s enough, there’s enough with the ITM data anyway. I’m sure <laugh>. So I guess, uh, let’s, let’s get into it. Uh, so I, TM for those that don’t know the acronym inspection, testing and maintenance, uh, data exchange, um, I guess, well, I guess what, what is it? And then we’ll get into why, why it’s, why it’s important. But, um, Bart, I guess, what, what is it, right? What is the, what’s the, what’s the point of it?
Bart van Leeuwen: (16:14):
Um, so, so I’ll, I’ll, I’ll give my perspective. I think Victoria should fill in at some point as well. Um, what we try to develop is, is a methodology to develop a, uh, data model, which, uh, supports the exchange of TM data. Um, and we are not, uh, creating the TM data model used for NFPA 25, but more a guideline on how you come to a point that you can create a data model, which is exchangeable, which is ex, uh, uh, extendable, uh, and which is usable by multiple parties to, to sort of integrate all these various data streams. Um, and in, instead of just, um, taking a sheet of paper and draw a data model, and this is what all the data of ITM should fit into, we really looked into what are the processes of how you create an I tm, uh, data model.
So if you, if you look at the, the, the 25 and the 72 standards, which are sort of pri primarily the targets of, for this project, you see there is some information about data collection, but nobody has ever thought about how do you exchange this information and, and how do you make sure that this can be exchanged and that data is reusable, uh, on various levels. Um, and, and that was more of a question that came up from previous projects, and I think that is probably the point where Victoria can more explain what led up to this stage where we are right now.
Victoria Hutchinson: (17:47):
Yeah, I, I think we need to provide a little bit of the context in terms of, of why we got to this point. So the foundation, we’ve done several projects. We’ve hosted several workshops on the topic of I T M data collection, um, and analysis. And really, it, it originated with, with a project, I think it was back in 2012, where they really wanted to look at, um, a fire pump performance. And, you know, when are we inspecting fire pumps on the right frequency? Are the, are the churn tests that we are doing weekly, is that actually wearing down the pump? Um, more than it’s, than it’s giving us benefit, right? So there’s a lot of questions, um, that they wanted to be answered. So we initiated a project where we tried to collect some, some data to try to try to come to a conclusion to evaluate what the frequency is currently in the code, is that appropriate?
Should it be changed, but have, you know, substantiation to that through, through real inspection testing and maintenance data. So, but you know, in that project we ran into challenges with, with the data collection, right? And anyone that you go to, um, to get this data, it looks differently. It’s collected in a different way, everyone doesn’t have, you know, all the same information. It was a very manual process in terms of, of collecting it and being able to, to analyze it. So in that project, we were essentially taking what we got and putting it into, uh, a separate Excel file and creating our own and trying to manually compile all of this. Yeah. But to be able to make a, an, a big decision like that, a big change in the code, that process, you would never be able to get the qu the quantity of data that you needed by doing it that way.
You weren’t gonna get the, the quality of, of information. So it really wasn’t a, a su sustainable process, but that just looked primarily at, at non flow, um, pump inspections. So then a couple years later, we did another, um, uh, we did another study that looked at, you know, can we predict when a pump is going to fail? Well, there’s algorithms and, and things to be able to do that, and we’ve, you know, developed what it would be if you had the data to support it. But again, the biggest challenge was the data and the fact that it was not in a consistent format. It couldn’t be analyzed or exchanged. Um, so that kind of brought us to this current need is, you know, everyone wanted, everyone wants the analytics, everyone wants the data to inform the standard or to inform their own practices. You know, there’s, there’s large corporations or facilities that, you know, they spend a lot of money, a lot of time, um, conducting these activities.
And a lot of times the, the data from their own facilities around the world is disparate. So they can’t even com compile, you know, to do a performance-based program for their own facility. So that, that was one of the requests as well. So, you know, instead of going onto another phase, there’s a lot of, you know, push continue to do more. This is a problem, that this is something that we really wanna do. Um, this is something that we wanna move forward on to get that analytics. Um, but we kind of had to take a step back because we said, okay, well the, the typical process that, that we’ve used is, is not going to work in this case. So we, we have to take a step back and address what the, what the problem is. Um, so that’s, that’s how we came up with this concept of developing a data model to basically get us on the same page of, um, you know, terminology for the data, what the, what the data means, how it’s connected, what the relationships are between it, what, you know, a format should be.
And again, you know, what we’re doing here is we wanted to, you know, try to see samples of, of data out there of how could we, you know, transform that and also, you know, kind of understand, so we can provide some, some best practices, um, in terms of terminology and, and format, um, so that we can get to a point to someday we could, you know, put out a request and say, you know, this is one question that we have that we we’d like to do some research on, is, um, you know, are people able to submit data, um, to, to research that? And that’s something, it’s not within the scope of, of this project, but it’s something that’s, that’s, you know, we’ve kind of hoped to be able to do that in, in the future. But to be able to do that in the future, we had to develop something that could, could help us get the data on the same page.
Drew Slocum: (22:32):
Sure, sure, sure. So, I, uh, a few things you hit there. Uh, this focused on N F FPA 25 and 72, the sprinkler I TM standard, and obviously the components of fire alarm as well within 72. Now, you, you mentioned the, the, the fire pump study back in 2012. Is, is, is that, is that the reason for the fire pump, no flow or churn test flipping from a, a, a weekly to a monthly and then back the other way? Was that, is that the cause from, is, is that all from this project?
Victoria Hutchinson: (23:09):
I’m honestly not sure, um, if that was, was the reason for that. I
Drew Slocum: (23:14):
Believe it went from weekly, I forgot if it was diesel, or I think it was, might have been diesel or electric. I forgot what pump it was, but it went from weekly to monthly and it might have went back as well. So, um, it was probably around the same time that the analysis was done. So, um,
Victoria Hutchinson: (23:30):
Yeah, I know that there was definitely discussion on, on the report at the, in the codes and standards process, but I’d have to look up to see if, if that was a substantiation for the change. I’m honestly not sure.
Drew Slocum: (23:42):
Have, have you been, have either of you been to an NFPA 25, uh, committee meeting before?
Victoria Hutchinson: (23:48):
I have <laugh>,
Drew Slocum: (23:50):
Nope. Why do you, why do you laugh? Victoria <laugh>.
Victoria Hutchinson: (23:53):
Oh, no reason
Drew Slocum: (23:55):
Bart van Leeuwen: (23:57):
So, so, so with all this, I’m starting to get curious.
Drew Slocum: (24:01):
Yeah. Oh, all right. So, uh, it’s funny why everybody, I, I bring, I haven’t brought that up in the podcast before, but, um, I, and right when, uh, I was going full-time with Inspect point, I, I, I attended one out in Arizona, and it was, it was eye-opening and of how committees work. Um, and that was really the first one I went to. And there’s just so many, I hopefully I don’t get in trouble for saying this, but there’s just so many different peop experts and professionals from the industry in the room. There were so many opinions and, um, and everything going into this meeting, I felt like the code couldn’t move forward fast enough. And it was just like, and, and, and I know there’s, there’s been changes since then to, to kind of, you know, expedite things. And, you know, you know, what actually helped, I, I’ve been some recent N F P A 25 committee meetings and the virtual aspect, uh, of it has helped tremendously because we can, we can meet quicker or, or meet more often and there’s not travel involved. So I, I actually think the virtual, uh, meeting that NFPA is kind of moving forward with is, might help the whole cold process, uh, out. So, um, yeah,
Victoria Hutchinson: (25:21):
I think it might also help more, more guests get involved too, right? Is not having to travel to meetings. It allows for, for more perspectives to, to come to the table.
Drew Slocum: (25:32):
Bart van Leeuwen: (25:33):
Yeah. Having, having been on a completely different industry on, on standardization fields, on, on data standards, I, I, I’ve seen that sort of the, the, the face-to-face meetings are absolutely helpful, uh, and, and help you build relationship. Um, it certainly helps. You don’t have to travel half halfway around the world if you just need to get stuff done. Uh, I, I think that will be something that would be interesting for the NFPA to, to look at. But I’m not saying that in any capacity. I’m just saying that <laugh>, that’s a bystander, basically.
Drew Slocum: (26:07):
Right, right, right. Um, now it’s, it’s all great. So I guess, what is the, uh, you kind of got into the why, why we’re doing this, um, and I guess you said what is the trouble right now? How, how long has the project been going on Victoria?
Victoria Hutchinson: (26:30):
Uh, I think it started in the summer of, of 2019.
Drew Slocum: (26:35):
Okay. So it’s, it’s still pretty, I, for some reason, I, I, I’d heard it’s been going on for a lot longer, but, um, but those were previous maybe with fire pump.
Victoria Hutchinson: (26:45):
Yeah. I mean, the previous projects, those were, I mean, the past couple years, I mean, I mean, some people see this as an extension of, you know, kind of like a, a phase three of the project. I, I, I don’t really see it that way. I kind of see it as not, not necessarily starting over, but it’s a, it’s a different focus than, than what the others were, were trying to, to accomplish. But yeah, it started back in, um, the summer of 2019. We hit a few, you know, bumps in the road when co D hit, you know, that was in the, the time period of when we were trying to, to collect data. And a lot of the people we were, we were trying to gather data from just, you know, had a lot of other priorities, obviously. Uh, so we kind of put a pause on, on things and allowed things to settle down. Uh, so that extended the project a little longer than we anticipated, but we’re, we’re close to wrapping up now.
Drew Slocum: (27:40):
So, uh, you mentioned, uh, um, a pro, uh, the problem of, of, of getting the data, and I’m, uh, I have a six Sigma, uh, background, um, within fire, fire protection, kind of, I was, I was doing it with Tyco, but, um, what, you know, I did a lot of projects over the years, and always, it was always the toughest part of the project was the, the measurement phase and, and grabbing the data, grabbing. Mm-hmm. Not even analyzing data. It’s grabbing the data. So can you talk to that a little bit about why it’s, why it’s been so difficult to grab the data and, um, you know, I can give my 2 cents on it as well, but, uh, I’d love to hear your perspective.
Victoria Hutchinson: (28:29):
I’ll, I’ll start and then I’ll let Bart chime in on, on his thoughts. Um, it’s data, data is sensitive. Um, you know, and there’s a lot of questions, especially, you know, in this sense we’re, we’re doing it for, for research purposes, but anytime you’re, you’re giving data to someone, there’s all these questions of, you know, what are you gonna do with it? What is your purpose? How long are you gonna keep it? Um, what are you gonna do with it after? Is there any way that this could, could come back and hurt me? Right? So I think there’s kind of an, a natural resistance, um, to giving data, especially when it’s something, you know, around the activities of, of itm. Um, and, and the fact that it’s required by the, by by the code and enforced by the hjs and stuff. I think there’s kind of a natural hesitation of, you know, well, what if we, what if we did something wrong?
Or what if there’s something that we missed and this is gonna backfire on us in, in some way? I think that, um, is a lot of it. There’s also, you know, for this project we had a, um, data sharing agreement in place. It was pretty straightforward, but we wanted to make sure that we clarified, you know, what the purpose was, um, what we were using it for, and things like that. And a lot of times it, it gets stuck, you know, in the, in the legal process as well. Um, sometimes, you know, if it, it depends also in terms of, you know, where their data sits, right? If, if someone has all paper based or, or PDF forms where they have to go and gather all of these to get you three years of data. Yeah. Um, that’s a really big lift from a time perspective, um, to get that if it’s, you know, if you have a, a software solution and you can, you know, download it and send you over a huge Excel file, it’s not so much of a lift, but that’s how everyone is collecting this data, um, you know, across the country and elsewhere.
It’s, it’s very diverse. Um, and it’s, it’s dependent on, it’s different for, for every party, um, that we’ve talked to. And then also there’s a, there’s another thing in terms of, you know, the ownership, um, discussion as well is, um, you know, the, the contractor may have it, but so does, so does the owner, and do you have, who has permission of it? Who has, you know, who has a copy of it and does everyone give permission to, to share it? So things get complicated there as well. But I don’t, I don’t know if Bart has any other Bob, you add.
Bart van Leeuwen: (31:12):
Yeah, I, I, I, I think this is, um, a fair share of the problems we ran into, but I think, um, um, if, if you go even back to, to, if you look at the, the, the, the codes and standards, the, the 72 and the 25, um, there is some wording that it needs to be transferable or exchangeable the results of the inspection. And there is a appendix, which gives you a guideline on how to record it. Um, but nothing sort of written in stone on how it needs to be done. Um, which basically means that even if the people were like, oh, we, we, we are willing to give it, there is no prescribed way of exchanging the information within the codes and standards. So that was something we have to had to sort of build up. So we created a sort of our own cloud instance, sort of Dropbox clone to, to be able to receive the data.
So these, all these elements of the, the, the problem of getting the data, um, is something we actually captured in, in, in the, in the document, in the report. And that’s about fair. Uh, it’s something you see a lot in the, in the, uh, research, uh, environment, which is called fair data. And that’s about findable, accessible, interoperable, and reusable data where findable is obviously, where can I find this information? And, and can I see who has a bit of information where I can request a specific that Victoria said, we wanna do a certain piece of research on a certain parameter in any of the standards, who has data that covers any of these parameters that’s findable, the accessible, how do I get to the data and how do I get the right, um, uh, the, the right credentials to be able to, to go to the data?
And what method do we use? And nothing like that, uh, is described in, in the 25 and the 72, and you’ll, you’ll see some wording in the, in the nine 50 to 9 51 on what could be the best practice, although they are more aimed at fire services, but this is more in the direction, how do you exchange data mm-hmm. <affirmative> then the interoperability part, it’s like, okay, and what is in the data and what does it mean? And we had several calls with people who very easily were capable of giving us the data, sending us large Excel sheets, and we were looking at the Excel sheets and we were, oh, my, I have no clue what this is, and why are some records double and what does that mean? And then you go over with the data specialist from the, from the, from the people who supply the data.
And they were like, we have no clue either. We just have to redo the export and clean it up for you. Um, so the interoperability was really hard, and we needed them to explain to us what certain things in these Excel sheets mean. And then the reusability part is, uh, not so much if you’re able to reuse it. And, and with that, I always use the example, it’s like if you find the picture on the web on, on, on the, on Google, and you say, oh, I wanna use that on my website or in my presentation, um, the reusability is not so much does this picture actually work in my presentation, but do you actually have the proper rights to use it? And, and that is obviously where the, the difficult part is. And, and, and everybody knows when legal gets involved, things slow down.
Um, so that is what we proposed as well within, within the document. If you, if you wanna go and solidify this before you even go and talk about how does the data look, you need to look at the fair principles at an industry where you say, well, we have a scheme in which we can exchange data with, for example, the research foundation to do research upon. And then we got licenses in place and everything so that every time you would like to do any of these, these type of research, you do not have to go and play the game all over again. How do you want it, where do you want it, how should it look? And what are the licenses? And what has legal to say about that? So, uh, that is an, an important thing that we’ve put in this document, which doesn’t say anything about the data per se, and if we describe it in the proper way, it could be used for any data exchange in any standard within the fire industry.
Drew Slocum: (35:16):
Sure, sure. Yeah. I mean, uh, the, the question I always go back to is, all right, you have contractors or facilities or whoever that, that have this data, right? And they have it, whether it’s digital, most of it’s digital these days, whether, you know, there’s still a lot of pen and paper, but what, what is the reason or what, why would they give it to the N F P A research foundation? What is their, you know, why would they actually do that? Right? I, I know there’s obviously a goal to improve the standard and all that, but at the end of the day, what, what, what’s in it for them almost, you know what I mean?
Victoria Hutchinson: (36:01):
I think the, the groups that have the most, um, direct benefit, um, from the data is really, you know, facility owners, you know, facility managers, right? Um, and insurers also have, uh, a big interest. And the primary, I think their primary objective in wanting this is being able to assess their own risk. Um, and being able to do, you know, kind weigh the, the cost versus versus safety discussion as well. Um, cuz there’s, there’s allowances in n FFA 25 that allow you to do a performance-based, um, ITM program. Uh, but you need data to be able to, to be able to support that. Um, so I think that, you know, and then also from insurance perspective, you know, if they can, if they can get data in terms of, uh, you know, better data in terms of like their in inspection record and the, the status of their equipment, you know, can they offer, you know, premiums or, or something or reductions for, you know, what they’re doing, um, in terms of their, their fire protection.
So I think those are, those are the groups, um, that kind of have the most benefit. But that’s where you also kinda, you know, hit, hit a roadblock is the, the contractors and stuff are the ones that are, that are doing, um, the inspections that have a lot of this data. Obviously a copy is made, um, to the building owners, but sometimes you get into that discussion of, of who owns that record, you know, if it’s, um, if it’s held with the, with the contractor and, you know, they don’t have, they may have a copy of it, but they may not have it in a format that, that we needed, like for this, for this project. So that’s, that’s definitely, that’s definitely been a barrier. Um, for what, for what, you know, this project specifically.
Drew Slocum: (38:01):
Yeah, I, I think, you know, yeah, it’s definitely the facility owner. And what about the A H J though? Like they have a lot of data with a lot of the, uh, new methods that they’re collecting deficiency data and all that. I, I would think there’d be a big benefit for jurisdictions, um, for that as well. And I think they have ownership of, of some of that data. Maybe they don’t, but
Victoria Hutchinson: (38:25):
It honestly depends on, depends on who you talk to. Sure. Um, but they, they do have interest in it because they’re, they’re looking, you know, from the aha perspective. Um, I mean, I think they’re really wanting to know like, where should I prioritize? Um, you know, my, my focus because there’s only, there’s only so much time and there’s a lot more places that they, they need to be going, um, then they have time for, right? So they definitely have, have an interest in this, in this data. A lot of times, you know, from one of the ones that we, um, were having discussions with, you know, a lot of what they had available to them was essentially, um, just a, a scanned in p d F document. Um, and that’s, that’s difficult to, to work with without, you know, having, doing it manually, right? So, yep. That’s another one of the barriers too. But again, again, it varies per jurisdiction in terms of what, what format they have their data and how, you know, do they have it, you know, on their own servers or is it just made accessible to them upon requests? It’s, it’s somewhat variable.
Drew Slocum: (39:39):
Yeah. It’s, uh, it, but you do, you know, I saw the article that came out. Um, I Bart shared it this morning on, on LinkedIn, which I’m glad he did cuz I got to read it. I didn’t even see it out there. Uh, but it, it’s very timely. It’s, it’s, I think, uh, N F P A interviewed you Victoria, of, of, of this. So we’ll put the, we’ll put the, um, uh, the link in the show notes and stuff, but, um, it sounds like you, you were able to collect some data out there. Uh, I, I guess what did you collect? Is it NFPA 25? Is it NFPA 72? And, uh, how did that go?
Victoria Hutchinson: (40:20):
Uh, maybe let, let Bart take this one. Um, it’s from both, it, it depended on, on who, who submitted the data? Val, let Bart take that.
Bart van Leeuwen: (40:30):
Yeah, so, so, so we, we, we got quite a substantial amount of data actually. Um, we, we aimed for, um, a lot of providers because we wanted to see how diverse the, the, the types of data that are out there were. And, and that is something that didn’t really pick up as good as we as we hoped it would. Um, so we got four providers giving us data in, in, in various formats, but we saw some similarities between the formats, uh, that went from a couple of thousand records. So I think in total, we, we got in, in the, in the demo system we created, we probably got 40,000 inspections or something, um, ranging from, from, um, the, the fire pumps, the sprinklers, the we risers to also the, the, the fire panels and everything. Um, and, and, and we ran into some of the, some of the issues that you, we we’ve heard about on some of the data is, is granular on different, different levels.
Um, some of the data has very complex cross references where you need really a company specialist to help you make an educated guess about what it actually means. Um, some only create, uh, contain some upper level information. Some were extremely detailed up to serial numbers of things. Um, and that was predominantly from, from the facilities who had a lot of data. And, and that was really interesting for us to work with that. Um, but you could clearly see there was no predefined structure of how to I, TM data should look. Um, it went from having basically the sample forms in the NFPA document, then created in data sheets to some of this stuff that is in NX f put in data form, but, and then extended with what they would need for their own software. So it wa it was literally all over the place.
Drew Slocum: (42:27):
Yeah, I I i I, I can imagine I, uh, as you know, I can, you know, with inspect point we’re, uh, it’s, it’s interesting to see what, uh, contractors and even facilities are doing, you know, um, currently, right. And even what they were doing before. So, um, yeah, I, I can’t imagine, uh, what formats we’re in, but 40,000 is a pretty good data set. Right. Um, so I guess what’s the, what’s the results? What’s the results of that looks like from the article here, it says there’s something being released in May. Is that, is it gonna be game changing? Is it just like a little tease of what could this be?
Victoria Hutchinson: (43:09):
Oh, what’s coming out in May will, um, be the, be the report for the project, kind of documenting what was done. Um, we’re still having discussions. I mean, BART has developed a, a full and, you know, functional data model, um, in terms of how we’re gonna release that. We’re still trying to, to, to figure out exactly what that looks like, um, how we wanna carry the, the project forward. Is, is there, you know, kind of an analysis step that, that we can take, um, based off of this. So that’s, that’s kind of the next step. There will be some, some guidance in that report though that, that hopefully the, the committee can use primarily in terms of, um, you know, the, the structure and, and how the, the data could look as kind of a recommendation. But Bart do you wanna talk about that a little bit more?
Bart van Leeuwen: (44:02):
Um, yeah, I, I, I think one of the aims of the project was not to create a data repository. That is some, some of the discussions we’ve had with externals as well. It’s like, oh, you’re gonna collect all this data and then publish that or do something with that. That was not the intention. The intention of collecting the data was, are we able to have a diverse set of data and then use the data model we create to, to create a uniforms exp um, sort of a uniform looking set of data out of the various providers. And that was something we were capable of doing, and we, we, we could demonstrate that. And on top of that, you will end up with people doing analysis, but we were, we will only be sort of facilitating the ability to do that. Um, and obviously giving some guidance, if you want to put this into the codes and standards, then how it should be done and, and what you should look at.
Um, because we, we’ve noticed that there was not really a structure in how the, the proposed data exports were formatted and some of the best practices out there in the world were not used. So we collected them. It’s like, if you wanna create a data standard, look at what has been done, take that into account, and on top of that, create something that fits 25. So it’s not that we’ve created the data model for 25 now. Um, I think that is important thing to say because we had some feedback that people were thinking that this was it, and this is mely a start. Yeah. Am I missing anything Victoria, in, in this?
Victoria Hutchinson: (45:34):
No, uh, I think that that pretty much covers it. Um, but in terms of what we’re, we’re providing is we’re trying to provide guidance in terms of, you know, getting a standardized export, right? Is, we know that there is, there’s a lot of vendors out there and, you know, I think that’s, it’s really great for, for the industry as, as a whole. But what we’re hoping to try to get people on the same page on is, you know, can we get on the same page in terms of how we export this data, um, that would, you know, basically allow us to, to do analysis in the future. That’s, that’s kind of the, the ultimate goal. So again, you know, what we’ve developed is based on the data that, that we’ve received. So it’s a, it’s definitely a, a good, a good starting point for sure. And it’s something that the, the committee can consider
Drew Slocum: (46:23):
Is, is it, um, you know, in the annex of NF p a 25, right now, they have, you know, a recommended dataset. Do you think that’ll be altered or some sort of, uh, do you think the, do you think N F P A 25, 20 th 2023 edition, we’ll take into account some of this coming out in May.
Victoria Hutchinson: (46:50):
I’ll defer to you, Bart. Do you, in terms of how you think, you know, what’s in the annex could potentially change?
Bart van Leeuwen: (46:58):
Um, I, I think it should if it’s gonna happen. That’s not up to me, that’s up to the committee. Oh, yeah, yeah, yeah. Um, yeah, I mean, <laugh>, I, I, I think, I think some of, some of the, the way that data is, is recorded, uh, and, and, and a try has been made before, but that should definitely follow more of the guidelines and, and look at what we’ve done in this project. Uh, but like I said, I, I think it should, I think it, it would be a smart thing to do, but then I’m not, uh, the one who’s mandate that I can, I can, I only make the advice basically. Sure,
Drew Slocum: (47:33):
Sure, sure. Um, uh, so if, if listeners out there, or contractors or facilities and AHJs even that are listening to this, uh, uh, if, if, if they wanna get involved in this and be a part of this project, is there, is there a way for them to do that?
Victoria Hutchinson: (47:53):
Absolutely. I would say to, to contact me, you know, directly. Um, we’re still, you know, considering this at, at N F P A in terms of, um, you know, how we can potentially continue this or, or carry it forward to a, to a next step. So, um, anyone that that’s interested, I, I would recommend, you know, reaching out to me and can, we can provide my, my email and, and get in touch, um, in terms of, of how people can, can participate going forward. Again, this, this project is, is coming to a close. So in, so in terms of the involvement on, on this project, um, we’re, we’re kind of wrapping that up, but we do expect to, to move forward, um, in that, that direction is, is kind of being defined now.
Drew Slocum: (48:39):
Nice, nice, nice. You know, uh, this whole conversation, you know, the word data, if, if anybody is, if there’s a drinking game, uh, being conducted, we said data probably about 150 times, so, you know, you’re quite a, I must have quite a, a good buzz by now, so, <laugh> anyway. Um, well, thanks for, you know, uh, I, thanks for kind of summarizing a lot of this. Again, we could probably talk another hour on it, but, um, yeah, be interested to see what happens once that’s released in, in a month here. And, uh, yeah, very excited to see how that will be brought into the code and just the reaction to the committee of, uh, what it looks like, right. So, yeah, absolutely. Um, so before we get outta here, I do something called, uh, the quick response round. Try to keep it light. We have a, you know, data’s always a, um, a dry topic sometimes, but, uh, so I, I’ve got a few quick response questions. Don’t need much of an answer, but, uh, I try to make fun with it. So, uh, um, Victoria, so where, where did you like living in Worcester or Stillwater? Better?
Victoria Hutchinson: (50:01):
Uh, probably Stillwater.
Drew Slocum: (50:03):
Yeah. Any reason?
Victoria Hutchinson: (50:06):
Drew Slocum: (50:08):
Better sports <laugh>, better sports. All right. Yeah, I agree.
Victoria Hutchinson: (50:13):
And not as cold.
Drew Slocum: (50:15):
And not as cold. Yeah. WPIs, you know, sports, you know, they’re, I went to R P I, so we beat ’em all the time. <laugh>, so they didn’t really have a good, uh, sports program.
Victoria Hutchinson: (50:25):
Yeah. Coming from a big golf school, it, uh, the sports there were a, were a little bit of a let down great school, but
Drew Slocum: (50:32):
Right. Yeah. Both big and fire protection. Uh, Bart, um, I’m a, you know, I went to school for industrial engineering. I have six Sigma black belt with Antico. What <laugh> nerd question here. What’s your favorite style of statistical analysis?
Bart van Leeuwen: (50:51):
Drew Slocum: (50:52):
Or type? Sorry,
Bart van Leeuwen: (50:57):
That is completely outside my field. Statistical analysis. That’s not what we do. <laugh>,
Drew Slocum: (51:04):
I don’t know if you’re looking at data sets and, and, and had, uh, had a favorite. My mine’s always chi squared, so, uh, everybody can laugh at me in the pod podcast, but that was always my favorite <laugh>,
Bart van Leeuwen: (51:15):
I, I would’ve referred this to, to, to, to nick, my data scientist. He is a, he’s the one who’s doing all that stuff for us. <laugh>.
Drew Slocum: (51:21):
I, uh, I, uh, I used to work with, I used to work at Viking, and, um, I’m gonna give him a shout out, uh, uh, Jeff Norton, who may have reached out to, to you, Victoria at some point. He, he was with Vivic mm-hmm. <affirmative>, you know, Viking and fire protection for, for a long time. And he just started his own, uh, data science company, um, which is really intriguing. Went from, you know, essentially running VP of marketing and, and a lot of different roles at Viking. And now he’s, you know, he went into data science and, uh, he’s loving it, but, uh, I don’t know if I could deal with it every day, you know? So, <laugh>,
Victoria Hutchinson: (51:56):
Yeah, you can definitely get into the weeds.
Bart van Leeuwen: (51:58):
Drew Slocum: (51:59):
Uh, Victoria, what was, what was your favorite, uh, style of clean agent or, uh, suppression, uh, style when you, when you had to do that?
Victoria Hutchinson: (52:10):
Um, I actually enjoyed designing like Novec 1230 systems, um, and, and sprinkler systems as, as well. Um, you know, i, I come from kinda my, my family’s in the fire protection business, so, um, you know, before I was even in high school, my grandfather was teaching me sprinkler design. So that’s, that’s always been, you know, kind, kind of special to me that, you know, even when I was going into school, I’d, you know, like people always joke that I was, was raised reading, you know, NFP 13 bedtime stories, but <laugh>,
Drew Slocum: (52:47):
Oh man, it wasn’t, oh man,
Victoria Hutchinson: (52:49):
It wasn’t quite that dry. Um, but I think, you know, I, I did sprinklers for longer. Um, so when I started doing Clean Agent, that was, that was something a little new. Um, I just, I found it, found it interesting. I’m, I’m sure you’ve probably seen, you know, the road show, the, the first time that I saw, you know, Novak, and, and you can, you know, throw your phone in it, it catches your, your attention, right? So, um, I, I enjoyed that, but I enjoy both, honestly. Um, the, and that’s, I think the joy of the fire protection industry, right? Is there’s lots of d different challenges you face, and the, the solution’s always, always different. So I don’t know that I have a, have a favorite, but I think, um, being presented with something new was, was interesting to me at the time.
Drew Slocum: (53:39):
Yeah, I, I went to, when I worked for Tyco, uh, Ansel, I, I was at one of their clean agent, uh, demonstrations, and they actually let us go in the room when, you know, after, uh, Novak 1230 went off and mm-hmm. <affirmative>, I probably wouldn’t do that today, uh, <laugh>, you know, but I, I, yeah, it’s kind of crazy that let, let us do it, but it, it, it, it weirdly enough, it changes your vocal chords and it it gives you, instead of a high pitch like helium, it gave you a very, very deep voice. Um, oh wow. Yeah. Kind of a funny story. So, very
Victoria Hutchinson: (54:16):
Well, I never, I never got to go in the room after I just
Drew Slocum: (54:18):
Pulled up. I, I wouldn’t, I would, you know, with all the chemical stuff, I dunno, I would never do that again, but, so Bart, what, what’s the favorite part? What’s your favorite part of being, yeah, you’re still a firefighter. What’s, what’s the, what’s your favorite part of, you know, day-today? Firefighting,
Bart van Leeuwen: (54:36):
Um, the, the, the unexpected part of it. Um, um, and, and especially, uh, in, in a, in a, in a big city like Amsterdam, although it’s not a metropol, but it’s still, um, a large city. Um, so yeah, you never know what you run into. And I, I, and one of my presentations, uh, I I refer to that, uh, my mentor when I joined this fire services, you will never have fire at normal people. And my response was, we have 170 nationalities. Every major religion is in our city being actively practiced. Every sexual preferences and all the fetishes that come with it is being practiced is in our city, and all the drugs mankind ever conceived or at least being consumed in our city. So yeah, what is normal <laugh>? That’s that. Yeah. Interesting. That is really, uh, and, and we run in, in, into all the problems that these people run into, so that makes it, yeah, I, I got some great offline stories to tell <laugh>.
Drew Slocum: (55:38):
Yeah, I’d love to hear some sometime. Yeah, don’t, totally,
Bart van Leeuwen: (55:41):
Yeah, they’re not suitable for PO podcasts, trust me. But
Drew Slocum: (55:43):
It makes it really, oh, my, I, my, my podcast listeners. Come on. You can, you can toss anything on. No, I’m kidding.
Bart van Leeuwen: (55:49):
Drew Slocum: (55:52):
Um, well this, this has been fun. You too. Uh, again, thanks for, thanks for coming on the podcast. Uh, you know, I’ll put a lot of show notes in here. So I guess where, where can we find you? I know you mentioned you, you know, your email Victoria, and how, how can more people get involved and, and in contact the research foundation and, and, and you as well, Bart?
Victoria Hutchinson: (56:16):
Yeah, so I would say in terms of getting in contact with the, with the research foundation, I can, you know, provide my email for you to, to put in the notes. There’s also a email that can get to us firstname.lastname@example.org. Um, if you have a research request, we have a, a portal for that. Um, you can also find me on, on LinkedIn and be reached in a, in a lot of ways, but, um, definitely looking forward to, to carrying this forward, and not just this topic, but, but other, you know, issues that are important to the industry.
Bart van Leeuwen: (56:49):
Yeah, same goes for me. Um, people can read me, reach me at Bart at, uh, Neth dol. I have a LinkedIn profile, although my name in Do in, in the Netherlands is not really unique, so it might be tough to search. So I, I think it should be in the comments or the, the notes of the podcast. And yeah, I’ll, I’ll, I’ll
Drew Slocum: (57:05):
Definitely toss it in there.
Bart van Leeuwen: (57:07):
Company’s webpage that you can find what we’re doing. So, uh, there’s plenty of reasons. Oh, and my Twitter handle is Semantic Fire, which is a sort of a mix between, uh, uh, firefighter pictures and, and data ramblings. Um, and I try to do most of it in English. Yeah, great. Some, some my frustrations are out of in Dutch, so then you don’t have to bother about it,
Drew Slocum: (57:27):
<laugh>. Oh, that’s great. Well, again, thanks again and Um, yeah, looking forward to seeing some of the results from this.
Bart van Leeuwen: (57:37):
Great. Thank you. Excellent. Thank, thanks for having me.
Drew Slocum: (57:40):
This has been episode 29 of the Fire Protection Podcast, powered by Inspect point. I’m really glad to have Victoria and Bart on the podcast today to discuss more about the Research Foundation. So appreciate them for coming on, and hopefully we hear more about what they have, uh, of what their results are in May. So, um, again, we’re gonna be releasing this in April, so it should be pretty timely. So again, thanks for everybody and please like, and subscribe. Take care.