“I see of lot of people buying apps before they bought the smart phone. So our smart building strategy is to first build the smartphone foundation."
The Nexus podcast (Apple | Spotify | YouTube | Other apps) is our chance to explore and learn with the brightest in our industry—together. The project is directly funded by listeners like you who have joined the Nexus Pro membership community.
You can join Nexus Pro to get a weekly-ish deep dive, access to the Nexus Vendor Landscape, and invites to exclusive events with a community of smart buildings nerds.
Episode 62 is a very enlightening conversation with Jon Clarke, Head of Smart Building Technology at Dexus, one of the largest real estate developers, owners, and operators in Australia.
We talked about Jon's long and successful career in smart building technology or building technology as he would call it, and what he's learned from seeing it from all sides, the contractor side, the consultant side, and now the customer side.
Then we took a bit of a deep dive into Dexus' smart buildings program, where it's headed. Lots of parallels with recent owner-side episodes with Google and Nuveen.
Without further ado, please enjoy Nexus Podcast Episode 62.
Mentions and Links
- Dexus (0:37)
- Google's plan for smart buildings at scale (37:42)
- Gateway Biometrics Touchless Technology (41:54)
- Honeywell (42:21)
You can find Jon Clarke on LinkedIn.
- Jon's experience as a contractor, consultant, and customer; how they're different and can work together (12:45)
- How and why the industry will shift from technology getting pushed on owners to owners doing the pushing. (21:33)
- Dexus' smart building strategy (26:01)
- What they've learned from 5+ years of FDD and where they're headed with it (29:33)
- 3 places the industry needs to innovate in the FDD process (39:47)
- Jon's experience setting up an entire touchless office building BEFORE the pandemic (45:32)
- Jon's strategy he calls convergence of projects (53:55)
- The state of the smart buildings industry as Jon sees it (57:19)
Note: transcript was created using an imperfect machine learning tool and lightly edited by a human (so you can get the gist). Please forgive errors!
James Dice: [00:00:00] Yeah, welcome to the nexus podcast. I'm your host James dice each week. I fire questions at the leaders of the smart buildings industry to try to figure out where we're headed and how we can get there faster without all the marketing fluff. I'm pushing my learning to the limit. And I'm so glad to have you here.
Episode 62 is a very enlightening conversation with John Clark, head of smart building technology at Dexus, one of the largest real estate developers, owners, and operators in Australia. We talked about John's long and successful career in smart building technology or building technology as he would call it.
And what he's learned from seeing it from all sides, the contractor side, the consultant side, and now the customer side. Then we took a bit of a deep dive. Dexuses smart buildings program where it's headed lots of parallels with recent owner side episodes with Google and Nuveen without further ado, please enjoy next podcast, episode 62.
Hello, John. Welcome to the next podcast. Can you introduce yourself? So I'm John Clark and I'm the head of smart building technology at Texas based in Sydney and Australia. Cool. And I'm excited to talk about all of your view, your vast experience in the industry today. Let's start with your background.
How'd you get to Dexus. Uh, and then you can start as early as you'd like.
Jon Clarke: [00:01:27] Oh, wow. Um, what a bit of a dinosaur, I think in the, in the industry started in the early eighties. So I've always been in, um, automation, instrumentation and integration, if you like. And it's pretty sophisticated, even back then to the process, I guess.
Um, and through just, uh, you know, a number of things through redundancy, I fed into buildings in kind of, probably around about the late eighties and back in the day, it was all very antelope, proprietary models. Vendor manufacturer driven. Actually, I have to say it's a bit of a culture shock because I've gone from systems that were very fast, quite sophisticated to systems that were quite slow.
And, and the thing that there was the three principles that I'd been training. Where about safety, obviously first reliability and then efficiency. And those kinds of things went out the window. When I got into building different, it's a completely different game. Right. So, yeah, so I was a service tech for, uh, for a few years, for a pretty independent company, and then started my own integration company as a subcontractor in, I think it was 91.
James Dice: [00:02:35] Okay. So that was on the industrial side or that was on the buildings
Jon Clarke: [00:02:40] on the building side. So I was only in industry for about five or six years. And then when it went to buildings, but that five, six years of training was unbelievable. Yeah. The way it makes you think vibrate differently. And I think you'd bring that with you throughout your whole career, you know, it's kind of it's in your mind there.
James Dice: [00:02:54] Yeah. Why do you think the industrial side is, or at least was back then ahead of things on the building side?
Jon Clarke: [00:03:03] It's the value of what you're doing. If you think about, if you're thinking about the value of oil and the safety aspects around a refinery versus building automation, you know, that they're worlds apart.
So they had the money and the time to spend on the systems. And again, about the reliability. If you had a plan that was offline for a couple of hours, you're talking a huge amount of money. You know, if you have a system that doesn't work very well, you've got life safety issues. So they're very different.
Totally. It all goes back to money, I think
James Dice: [00:03:34] fortunately or unfortunately. Um, okay. So you had your own integration firm and then what was after that?
Jon Clarke: [00:03:41] I was really lucky. I, I started working for one of the two that I've mentioned the manufacturers who are just starting out on to the sequencing. Energy efficient to the controllers.
Cool. Okay. So I started doing some very early integration for those guys, with our company. There was a company a bit like, um, tritium in the UK that used to do integration drivers, cat. So we were, we would have helped the company integrate their chillers into building management systems and all that sort of stuff.
So I did that for quite a number of years, got to see some, um, really quite amazing projects, um, pretty random ones where. Chili's required and then.com came along in the, in the mid nineties. And I pretty much was doing data center designs for yeah. From probably about 94 through to about the early two thousands.
Okay. Everything was data science driven and obviously the cheating refrigeration requirements of a data sensor is pretty phenomenal. So yeah, that was my world for quite a few years. And that was, uh, an amazing learning curve. And it's not just about the HVAC systems. It took me back to my industrial process because it is almost like a process has to be reliable and resilient.
You know, you can't let these things fail. And then once you start looking into white space, you see where the computer its system systems, you really get a bit of an insight into what's happening in the it world at that time. So that's when it really started the OT it bit started to come together. I think in that sort of like midnight.
James Dice: [00:05:04] Okay. All right. Do you think it was coming together in your mind or was actually coming together in the industry as well?
Jon Clarke: [00:05:11] I think for those that were involved in data centers, they probably saw it. But for those that were just in HVAC and buildings, probably not.
James Dice: [00:05:17] Yeah. Yeah. Interesting. All right then what was next?
Jon Clarke: [00:05:22] I became the technical director of a larger company for seven years, got involved in some really great projects, some pretty crazy ones as well. And, um, and then I realized what I was doing is I'm very much about. Practicality execution value. And I was spending a lot of my time doing things that were not really techie.
They were more corporate focused looking at finances. And I had a lot of staff and, you know, so I was like doing company stuff rather than the back to your tech staff. So I started to kind of get a bit of fate and I wanted to go back into the realms of getting a bit more into the tech again. Yeah. And then in 2008, very early on, off duty came up to immigrate over to Australia from London.
And, um, I was a consultant until two, this role. Yeah, until 20, 20, 19 29. Cool. I'll say three sides, right? Have liked being integrator contractor consultant. And now,
James Dice: [00:06:19] all right, two questions. You mentioned crazy projects. I need some sort of example. Uh, what's the craziest project you did?
Jon Clarke: [00:06:26] British American tobacco is a tobacco drying plant in the middle of the Southern Russia.
Desert is in a place called is Becky style. Actually, it's funny enough. I talked about it on one of my last few days when I was with. And I found it on, on the, on the web. So I just talking to the engineers about what I did on that job. And it's still, you can still see it, but it was, um, so it was process control.
Obviously it was, you know, looking at drawing tobacco, but it was just the environment. I mean, you know, you flew to Tashkent, you're about six hours in a four by four going across the desert to get to the site. There was, uh, some pretty crazy individuals around you getting you there and making sure that you were safe.
Yeah. Yeah. And then, um, and then the actual process itself, not that I'm an advocate of smoking, but it's very, very complex that tobacco industry. And so that to get the humidity levels, right. And to get the tobacco right. To roll it into the cigarettes is a phenomenal kind of a process why it was so interesting was that we built package plant rooms in France, in a factory in France, because they were so big.
They couldn't get them on the UK road. To then get them to the port to get them to Europe. So there was a factory hired in France. We were there for probably about three months building these massive PLA plant rooms. And then they all got, um, taken by truck down to the site. We thought that was the end of it until we got a call probably about six or seven weeks later saying that they'd be damaged and we just need to send a team in to see what the damage was like.
Okay. And the damage was, we almost had to all the things again on site. So I had another crew go out for about 12 weeks. To try and rebuild the damaged system. Wow. Imagine, um, I'm from, you know, the almost middle of nowhere in the Midwestern United States, I've been to some really cool buildings, but that, that sounds pretty.
Yeah. All right. Um, my other question was. Well, a little bit more detail on why, why you left, why he left London. I've traveled a lot. I've been really lucky that, you know, my first job I traveled quite a bit. I've always kind of like loved going to new countries and finding new things. And I got relations in Australia over here and visited and I always kind of liked what it had to offer.
But it was a long way away. Yeah. I think the thing that got me was that it was what's next for me. And I have this thing, whereas I don't like to be comfortable in my comfort zone for too long. I always love it to try and be slightly on the edge of my comfort zone. I think it's the only way that you can kind of really progressing push yourself.
Yeah. So when an opportunity came up, I just decided why not? And, um, my family and myself moved out here and haven't looked back it's it's home for us. It's we love it. And I think at the time, you know, the industry, I was probably pretty advanced in what we were doing compared to what industry was doing here.
So I was bringing quite a lot of knowledge into, into the industry. And I remember one of the first projects I did actually for, for Dexus. We to optimize one of their buildings and the original plan was to rip out the BMS and, um, some other works, but I kind of flipped it on its head and said, no, the BMS still got a life left in it.
Let's rewrite all the software algorithms. And I rewatched it on a, on a, an algorithm. That's actually got like a voting pattern in it. And you haven't heard of it before? I didn't, I didn't, um, created it sits, come from another manufacturer, but, um, I, I took the best. It's um, the Truman response principles.
Yeah. And the neighbors rating from that building went from, you know, two and a half star to 4.7 and then they then upgraded the chillers and you've got it to 5.2, but the, you know, the, the big piece of the, um, the action was actually in rewriting the software algorithms. Cool. Yeah. And, and it's great to have, you know, they were very progressive Xs, love to try things out and they just said, You're you're the occupant of the building at the time.
It was where our office was even ago and it see what we can do with this thing and the results speaks for itself. So, yeah, really cool. And so what made you go from consulting to working at, at Dexus? What was that transition? Like? I think it was a natural progression for me. I got to the point where I kind of done consulting long enough from I realized I wants to do something else.
Yeah. Going back into either vendor or manufacturing world was kind of like going a bit backwards. So I think it was naturally the client role. I think it was a good fit for me. What I would say is having spanned these three major parts of the industry has given me a really good lens to understand the agendas and the complexities behind each of those.
Um, so yeah, I just like when I was a contractor, I used to get a bit frustrated by some of the consultants decisions. That's because I didn't realize what was happening in their world. And as a consultant, the same sort of thing with clients, making decisions you think, well, I'm not sure, but that's because again. Now I've seen all three
I can see the whole picture it all makes sense. And it is it's, it's a really challenging industry, I have to say. A lot of contradictions and a lot of agendas and things that, you know, a lot of moving parts.
James Dice: [00:11:37] Yeah.
Jon Clarke: [00:11:38] So I guess now it's going to be the ability to work out. What's the best execution plan for if we're going to roll out a tech in a building. I understand I've got to get a contractor involved and a consultant, and I've got a vision, how do I bring those three worlds together
and actually realize it?
James Dice: [00:11:53] Alright, this is really rare that someone has these three perspectives. What would you say the different mindsets of the three are and how, how do they differ? And then the second question is like, how do you, how do you bring those together, if there's a short answer?
Jon Clarke: [00:12:09] Um, okay. So I it's quite aggressive marketplace, right?
Particularly the industry for bidding for work. So the contractor really, if he wants to win the work, they're going to do exactly what it says in the brief, if they can. They need to be very careful if they offer more than the brief for obvious reasons. Yeah. So in their world, it's about what am I being asked to do?
How can I price it up most efficiently and deliver upon it? And that's literally in a box. Yeah. In the consultant's mind. It's again about, they've been given something to design, but they're designing against a timeline and a budget. Right. So again, what can I do for my client in this time and in this budget?
Yeah. And so that's how they kind of, it's almost like, you know, they're passing the problems they have to the contractor. And from the client's perspective, there's a big difference between an existing building retrofit and a new development, completely different process. In some aspects, a retrofit upgrade is easier to get the best results than a development project, because there are less moving parts and contractual obligations.
There are less people in chain. You know, I want you to replace a chiller, consultant inspects the chiller, contractor puts the chiller in, very very simplistic. But once you start to get into that development world, it's well, we have a brief, we have a vision, this is what we want a builder to build for us. But there's a lot of parts in that about, you know, um, getting a design consultant, to do a design to a certain certain level.
And then they hand the design over to the builder that's going to finish it. And if you think about the guys that are building it, they've got to price it, again on a time, you know, a timeframe, right? Because it's very, it's imperative that that building is occupied by a certain time. Yeah. The really big problem comes in is when you start to look at the technology or the bolt on parts. Pouring the concrete and everything else is one thing,
but once you start to go into technology and you've got a building, that's not going to come out of the ground five years, how do you set your price now to win a job on something that may not be developed in five years time? That's, that's the major challenge and that's an industry challenge that we're all working through at the moment.
And the only way you can overcome that is by changing the process. It's not about the tech, it's about the process, it's about the people. So I think it's really important, but I guess having the three lenses helps make to put that together because I've got some IP on how those three operate.
You just mentioned that the brief will be very, very different.
James Dice: [00:14:40] The ones that you guys write versus...
Jon Clarke: [00:14:42] Well it's the brief, and it's also the methodology behind it. And that's a problem too, because if we can't tell the builder how to build. Yeah. So again, it's yeah, it's putting together a very comprehensive strategy.
So I'm working through, you know, um, delivery strategies, procurement strategies, um, new roles that are coming into the industry.
James Dice: [00:15:04] , well, yeah, I'd love to talk about that strategy a little bit more. Um, can you first talk about like, what, what is Dexus what's the portfolio? How many buildings, what types of buildings?
All that for people that don't.
Jon Clarke: [00:15:16] Yeah, sure. So look, we're one of Australia's leading real estate groups. Um, we, we, we manage and we own a pretty large portfolio. Um, we are Australia globally, by the way. Um, so we know our else, um, the own, I think it's about 15 and a half billion of office industrial and healthcare.
So we've recently gone into healthcare property. So we have quite a diverse asset class and, um, and we manage about 21 bids. Um, for third parties as well. So that's a pretty big, pretty big organization across, you know, owned and operated. And then we've got a development pipeline and you've probably seen some stuff in the news recently as to things that we're doing about 11 billion that we've got pipeline at the moment.
It's a pretty significant, yeah. Okay. And then we're in it for the long-term we don't kind of really build and sell. We kind of built to keep got it.
James Dice: [00:16:06] And then what's, what's your role? And then what does like the smart buildings team look like around them?
Jon Clarke: [00:16:12] I, I look at the strategy of how we rolled out building technology.
Let's just park the smart piece for a minute. It's actually building technology and that could be anything from a building management system, access control, you name it. So I've got a, I've got a piece to play in how we procure and deploy those technology. Okay. Team around me made up of I've got guided some come from the it side of the industry.
Okay. So he's got a really good handle on networking and security and connectivity. And his role is actually in the its fundamental role is in the, what I call technology infrastructure. Okay. He looks at communications network telecommunications, um, in building mobile coverage for that side. Then I've got a technology.
And his role is that we get so many vendors approaching us with the flavor of the day. Yep. So he is the gatekeeper. If you like, he, um, we have a bit of a, uh, way of, um, registering what we look at, whether it's got a use case with regards to the next level of actually seeing a demo or whatever. So he does a lot of that and he also helps out with some of our project stuff as well.
Cool. And then I focus more on the kind of I'm across a bit of everything really, but then we've also got a project delivery group. So when projects go live, they help us with the project management side of side of it. Um, so there's quite a lot of internal internal stakeholders and there's a, there's a big internal operations to be able to help us deploy projects, which is, which is good.
And then the team, I guess, sit in the center of operations and sustainability, you know, I mean, we work very closely with the sustainability guys to help them with their targets and to, to enlighten them on which tech could help them out and how, yeah. And then we also sit across the operations guys who are looking at doing refurbs and retrofits to help them out with like an internal consulting group, I guess.
James Dice: [00:18:09] Yeah. It's interesting that you have an it background on your team. Can you talk a little bit about that? Why that is specifically, but also there's something I forgot to ask you about which your your shift from like this OT-centric world into this IT-centric world, you said that came from one of your big consulting clients.
Can you, can you elaborate more deeply on that?
Jon Clarke: [00:18:32] I mean, when I was an integrator, you could see the IT world starting to come into what we were doing. We saw some very early adoptions of IP networking. It was pretty primitive, but it was IP. And so you can kind of see the way the interview was going and it was more to do with about, so it was communications and remote connectivity.
You can see, they were the main things that were coming in to the industry. And then I think the product manufacturers were also starting to look at changing up their hardware to be more IP compliant, faster and more powerful systems. But one of the last, the gigs that I had with NDY was consulting to a big, big tech tech giant.
And my role there was to help them understand building datasets, what they could get out of buildings. And I remember having conversations where I was telling them about, you know, BacNET, how buildings worked, and built BMS systems and they were a bit shocked to be honest. They were like, really?
Are you sure? But you know, the other side of that hearing about their world and understanding where they were coming from, it blew my mind. It was such a different world and it's, it really opens you up to so much more power, scalability, um, ease of management, there's a phenomenal amount of benefits from embedding good
IT practices into your, into the OT world at that enterprise layer. Yeah, that was like three years where I was a sponge. I was just sucking up every minute I could of what was going on in their world. It was, it was fascinating. Yeah.
James Dice: [00:20:14] I think you and I maybe think similarly along the lines of like other building owners and operators are gonna go in that direction as well.
What do you think that's going to do to the industry?
Jon Clarke: [00:20:24] I think that what's going to happen is the demand is going to be reversed. At the moment, the demand has been vendor-driven. If you, if you know, you've only got to get somebody that goes on to Google or typed in smart buildings and you just see all of the amazing images that you get coming up in the cool tech and everything else. And they've all got some sort of a, um, a problem to solve. And I think that for a lot of building owners, they're not pretty sure they have those problems yet. So it's very much industry driven that I've got something that's going to help you out.
Yeah. And I think what's happening now, I think that from our perspective, we're seeing the power that data can provide a corporate level. And we're starting to realize actually we need to be able to also use the data from buildings that we don't have at the moment. So I think that the building owners and operators are going to start now demand new things.
So I think those two worlds are going to come together. I think the vendors are going to start to create products that are more beneficial to the end user than where they think their products are beneficial. There's a lot of stuff out there that really is, you know, probably not gonna help us out too much at the moment.
James Dice: [00:21:33] I'll everyone knows my opinions on that. So that that's, I agree with you. I think that's where we're headed as well. And I can't honestly can't wait for it. Um, you know, when I talked to the similar tech minds that come from the ITU world, I'm like, can you please keep educating people? So let's talk about the strategy of Texas, what what's, uh, what's like unique about your guys's approach and kind of what is the overall smart buildings, right.
Jon Clarke: [00:22:00] Well, I think I've indicated that we kind of don't buy shiny things unless they've got a really, really good value. It's about value proposition, right? Everything that we do is underpinned by minimum value proposition. Okay. So my strategy is more around building a foundation to digitize the portfolio.
Okay. Um, so it's about connectivity, uh, networking, good standards of cyber, trying to stay out of proprietary world, looking at open source and getting to a point where we've got a digitized portfolio that we can start to then gain some real insights and some power from. Okay. I kind of seen a lot of people buying apps before they bought a smartphone.
So it's about building up that smartphone foundation. Yeah. One of the first things I did was to start writing our technology standards for all of our primary systems. Okay. And that was to create consistency. And also to then to look at, does that open up opportunities for economies of scale, procurement strategies?
Yeah. Yeah. So there's a lot of stuff that we're doing is kind of like in the back end at the moment. I'm preparing for what's coming for the future and that's because I don't think really everybody, I don't think the industry is ready to really start to invest in a huge portfolio thing at the moment, because it's fast moving.
And I think that if we did invest in something, we might not get the power out of the product because of maybe the data sets aren't ready. Totally. I got to remind the learning of that, that as I said, working for that, those tech guys, my learning is about it's all about instructor and data. Without that, nothing else can happen.
James Dice: [00:23:46] And have you guys gone down the road of the, sort of, like you said, point solution app before. I love that analogy. You bought the app before the smartphone, have you guys gone down the app point solutions sort of, uh, Mindset or like with pilot projects and then realize that it wasn't scalable or
Jon Clarke: [00:24:04] we've got a couple of apps for years, but they they're third party vendor apps.
And to be honest with you, the customers love them. It's not doing things like, you know, you can adjust your air conditioning and all that sort of stuff. It's more of a community app that brings customers together and it's messaging, you know, they, they, it's all about we, we've got a team that obviously, um, look at customer feedback.
We're very, very, I'm big on talking to our customers. What they would like and what they're seeing a, feedback's been where there's a building with an app. Yeah. They say it's great. It's, it's a way of us connecting. With, you know, things that are happening around the building and in the building. So things like, you know, and events and things like that.
Um, so yeah, we, we've got a couple of apps that we use and we've also got our, um, we've got a big instance of sky spot running across the portfolio, which helps us to keep up with our foot protection optimization. Cool. That's something that we rolled out quite a few years ago. Cool. And is that across the whole portfolio that you're doing that primarily office.
Okay. We've got a bit of folk protection in retail, it's different products, and we've got a couple of different models across the portfolio. So it's the same platform, but we're using two different vendors. And the reason behind that was that they have different approaches and for us, we were trying to work out
is there one approach that does everything or do we need to two different approaches to give us what we want? I did that when I was consulting and Dexus were a client.
. Um, and so about five to six years ago. And we're at a stage now where we're just looking at, okay so what does V2 look like?
And we're just knocking on that journey.
James Dice: [00:25:34] All right. I'd love to hear what these first five years were like? How have you integrated the analytics into workflows to get stuff done? Um, and then have you used a centralized model or has it been more distributed?
Have you leaned on the vendor? How have you sort of implemented the FDD?
Jon Clarke: [00:25:50] So the first
thing that I had to do is work out what was in the building is actually, could it provide information to the FDD in the first place? There was a bunch of surveys that went on and out the back of that across the two vendors, we had to have two or three different mechanisms to get information out just because the legacy systems were in the building.
James Dice: [00:26:07] That speaks to your strategy on getting the infrastructure, right first.
Jon Clarke: [00:26:10] Absolutely. Yeah. So yeah.so we had a two or three mechanisms, you know, some were, we were fortunate that we could just plug a box in and get the information out. Others were using emails, you know, CSVs and all sorts of things,
anyway we could get the information. So that was the first challenge. Um, and then pretty much when it was down to the vendors to deploy it back then, um, those guys went in and did what they needed to do in the buildings. They got the platform running. And stage one was just to get full protection operating and see what it could do for us.
Okay. Now we started collecting the stats and because like I was saying to you previously, we weren't really sure what the value, we thought he had value. Yeah. But we were a very early adopter back then, I think we were the first in Australia to do it. And so we, we've kind of been a bit cautious about, you know, um, does this thing really have value?
Okay. Absolutely. Absolutely. It does. It's like, you know, it's, it's probably one of the, it should be now a standard tool set for protection. So it goes into our vendors environments, and then one mechanism is that they both report back to our ops teams about, um, folks and and, uh, what they found that they can optimize. It's helped us out with our maintenance.
And it's helped us out with some forecasting as well, because of it's starting to tell us, you know, we have some problems with some equipment, but I'm going to go away. And when you do a replacement and one of the hardest things for us is our, is our forecasting.
James Dice: [00:27:32] Cool. Okay. Is there, I've heard people talk about this value.
Do you guys, is there like a business case to do that? Do you actually save money by, by more intelligent planning?
My gut is
Jon Clarke: [00:27:43] that you do, it's really hard to quantify, but things that you can really put your, you can put money to is when just an example, if fault detection picks up that something's running when it shouldn't, maybe it's running all weekend or a long weekend, because of maybe either through a fault or through manual intervention
or somebody's forgotten to take it out of manual, there is a direct relation to a cost impact and energy impact to that. So when you, when it's related to energy, it's very easy to kind of quantify. When it's other things it's a bit harder to quantify.
James Dice: [00:28:15] Very cool. What's the next version 2 look like? Or what's the, what's the strategy behind the next round?
Jon Clarke: [00:28:23] The fault detection analytics at the moment is very focused on if you like HVAC or building within the building box. We're looking at, you need to start combining those data sets with other corporate operational information. So I think it's proper portfolio analytics that brings numerous data sets into it rather than just as our building
and then there's our corporate.
James Dice: [00:28:47] So inside the building, are you expanding beyond HVAC too? And then when you get up to the enterprise level, what other data streams are you talking about?
Jon Clarke: [00:28:55] Access, control, lifts, um, you name it. I think that everything in a building, that's got the ability to provide information
we need to look at it and there will be certain data sets we'd extract, not all data sets though. I think it's very easy just to say we'll have everything, but I think we need to realize that when you start pushing data to the cloud, there's also another cost that kicks in. Storage has become much more affordable these days.
But with the use of APIs and starting to draw on the data sets, once you start to ask better questions, that's when the, the taxi clock starts to tick, right? So we have to be mindful of the operating cost of the technology, not just the, the CapEx cost of putting it in.
James Dice: [00:29:35] Yeah, I hear, I hear people talk about that sometimes.
And I hear other ones saying, oh, the cloud has democratized the cloud so cheap. And it's like, well, sort of.
Cool. And, and what do you think the business case will be then to take, basically take analytics up to that next level? What will that mean for the organization?
Jon Clarke: [00:29:58] the business case is kind of there because we'd been on our corporate site, we've been playing with data sets for quite a long time. We've got some very, very smart cookies, working in Dexus, very, very smart, and they are able to extract all sorts of information.
So I think that they already understand data really is the central their world at the moment, to help them with their operations, whether that be financial, leasing, whatever, doesn't really matter. It's all about information. So I think that bringing our buildings up to that level as well, it's just a natural thing that needs to happen.
And I think that the business case is almost like it makes sense to do that. So I think that the particularly when you start to grab the efficiencies that we can gain, you know, bringing in better procurement processes, economy of scale, forecasting, running things on more of a reliability index rather than a monthly visit index.
I think that would go towards, I did a project quite a few years ago using a Monte Carlo simulation, which was looking at reliability, indexing of equipment. And the theory was if you've got a bit of equipment that can run for a thousand hours without any problems, don't go maintain it. 950 hours sort of thing, no? Very high level, you see where I'm coming from? There's another one where they were saying it's actually cheaper to buy standby piece of equipment, six months before you think it's going to fail, then go and maintain it. Now very progressive, very out there, but you kind of understand the theory behind it. Yeah.
James Dice: [00:31:24] So the ability to start rethinking the way that maintenance happens across the whole portfolio is kind of where you're headed?
Jon Clarke: [00:31:31] Yeah, absolutely. Yeah. I'm literally using the data for the maintenance and intelligent maintenance, I would call it rather than just your, your routine. An interesting piece about that is that we've looked at how can we reduce maintenance? And there was some things that you cannot reduce having truck call for because there's compliance.
We went through a whole matrix of what are the activities that happens during maintenance, and there's a number of them where you have to send somebody to site for compliance issues.
James Dice: [00:31:59] Like what would be an example?
Jon Clarke: [00:32:01] Uh, looking at things like anything to do with water, pipe, condensers, you know, things like that.
And then it's like, if you've got to send someone to site, you can't send him to site just to do a five minute exercise of inspecting something and tagging it. So while they're there, they might as well do other things and those other things, would've been the things that you would have said they don't need to go on site and do. So it's a real balancing.
James Dice: [00:32:20] Yeah, yeah. Yeah. I heard a couple of thought-leaders nothing published yet, but they are telling me about how they're thinking about this. Sometimes I've heard, you know, do your preventative maintenance, but while you're there, fix those three or four or five faults that popped up, like things like that. But I don't think, I don't think we've gotten to a point where there is an accepted or like future state that we're all trying to get to, in that respect. Um, from my perspective is we're like let's analyze data, let's find faults, and we're not at a point as an industry where we're like, we know exactly what to do with the faults, once we get them. Is that, is that kind of how you're seeing things too?
Jon Clarke: [00:32:58] Yeah. And I remember when we first put the, um, the fault detection in, we were watching the amount of folks to be found and the, the OPEX cost to fix the folks.
And you're in like a bit of a bell curve for awhile, but this thing is finding things that we didn't know were a problem and they do need to be fixed. And until things settle down, you you've got a bit of an uphill battle for it for the first few months of going live. Yeah. Hey guys, just another quick note from our sponsor nexus labs.
And then we'll get back to the show. This episode is brought to you by nexus foundations, our introductory course on the smart buildings industry. If you're new to the industry, this course is for you. If you're an industry vet, but want to understand how technology is changing things. This course is also for you.
The alumni are raving about the content, which they say pulls it all together, and they also love getting to meet the other students on the weekly zoom calls in the private chat. You can find out more about the firstname.lastname@example.org labs.online, all right. Back to the interview.
James Dice: [00:33:54] How do you think we get to that point as an industry where, when we have a list of faults and we have like streamlined processes and automations to get them into workflows, such as, you know, the O&M staff, you know, external service providers, how do you, where do you see that going?
Jon Clarke: [00:34:11] I think that will come into it. I think that if you look at automation in factories and process plants, they've got that pretty much down now. I think the technology is there, I just think that people have certain processes in their organizations and it's a bit of a difficult shift for them.
It's a mindset thing. Yeah. I've, I've seen automated work orders before in other industries. I think the thing that's missing from the analytics is the, it's what I call the fact data. So the fact data is about, it's the intellect of the persons looking at it. So a machine may say to you, or I've looked at this and I've looked at that and I think the problem is here, but a human may look at it and go, well, actually, I can understand your thinking, but my experience, the problem is actually over here, and that's the fact data, right?
So I think there's still so much human triaging that happens in fault detection, it's very difficult to automate the workflow because you could be witted out. So I think it's still until you can get that machine learning to be very very accurate. I think that's when you'll really start to see the full workflow automation happening.
James Dice: [00:35:15] Yeah. This is a, this is a place where I feel like there's two types of innovation that's needed. One is with the data you have come up with the best root cause that's possible with the data, right. I feel like there's a lot of analytics firms that only go a portion of the way on that spectrum, right.
They're going to give you kind of a shitty piece of analysis, in my opinion. They're like here, human, deal with this, right. But then there's some analytics providers that gets you further along, right. But I feel like the next piece, which isn't like white space, as far as I know, I haven't seen any vendors doing this. I'd be happy to get about 20 emails after this gets published of all the people that are doing this. But what I haven't seen yet is when I give the human, the analytics and then they go make a decision, there's a shit ton of data there that happens, right. Um, they acknowledge it or they delete it or they go change the rule that came up with it because it wasn't right. They go dig into the controls drawings to figure out, you know, like well actually you're tagging or you're modeling was wrong, and then you're going into the implementation, right. Uh, that was a loose set screw, or that actually, you know, that piece of equipment doesn't even have a valve. You get into all that stuff,
right. And I feel like there's a ton of data there that we don't have that's currently not digital that we could be using. And I feel like that's like the next frontier to making actions more reliable or recommended actions more reliable.
Jon Clarke: [00:36:39] Yeah, and it's a problem because of, I guess it goes back to standards again. I mean, one of the things why I wanted to get these standards written, was that if I go to 10 consultants, I'll get 10 flavors of controls for my HVAC system. And those HVAC systems could all be VAV, it could be TOB, doesn't matter what it is, but I will have variations upon a theme. And yet HVAC is all based upon physics, right?
You can't change physics and psychometrics. So the fundamentals of HVAC are all set on some standards. There are these opinionated control strategies that tweak, and some of them are agree to really good. This is no disrespect to anybody in industry, but some of them are great. Some of them not so great, but they're just different.
Yeah. So now you went to try and apply some standard rule sets. It becomes a challenge because you almost have to reinvent the wheel and every time to understand the control algorithms that are working within the building and then adapt your rule sets to make sense of them. Yeah. Super interesting.
James Dice: [00:37:37] That's the third frontier that comes first. And I think when I had, um, Trevor and Keith from Google on the podcast last winter, they talked about creating those sorts of standards in terms of like, we will only accept these types of HVAC systems in our buildings and these control sequences, you know, that, that comes upstream to the FDD, right.
Yeah, absolutely, totally agree.
Jon Clarke: [00:38:02] Yeah. Yeah. So we we've written within our briefs now of how we want things controlled to try and standardize on that. Because when the really the new world really powerful AI comes along, we don't want to have to have very complex implementation at every single building, because obviously the business case is harder to justify with the amount of cost to deployment.
James Dice: [00:38:21] Well, I challenge the industry again, if anyone's doing AI on the implementation of the fault data, the interaction between humans and the investigation that goes into it, reach out to me. I want to hear about it. All right. What about like non FDD types of technology deployments? Um, I know you guys are doing, you know, obviously just announced this at LaSeon headquarters project, but then you've also done several other kinds of smart building, prototypes, pilots, whatever you want to call them as well.
So what have you learned from, from those projects? I mean, can you give some stories there?
Jon Clarke: [00:38:55] I think probably a good one to talk about is the gateway building where we integrated. The hand scanning biometrics into the access control. I think there's seven technologies that we have to pull together in that building.
Cool. And I'll just jump to the results, first of all, great project. It took a lot of organizing and a lot of collaboration between the whole team between the integrator and all the other vendors involved in the project. It proves that you can do it, you can get a great outcome, but a complex journey a very detailed, complex journey.
So with that building, it was going through a lobby refurbishment to the gateway building in Sydney. One of the customers in there was security conscious and wanted speed gates to go in. And as you know, once you put speed gates in that impacts on everybody, all other customers, visitors. So we have to think, okay, we'll put the gates in, but can we make this a nicer experience for everybody? What could we do to get round the typical kind of get your access card out and tap and everything else? Because it was a lobby upgrade, we were also refurbishing the lifts. So it was an opportunity to converge several projects into one technology.
And actually, there were some efficiencies in doing that, which is one of our strategies going forward as well. Um, which I can talk to about in a minute for like, it's about the convergence of projects to create efficiencies. So the gateway, um, yeah, we we'd already trialed the, the hand scanners and the feedback was pretty good on one of our buildings.
So we knew that adoption would be okay. Just on that, we were very unsure about using cameras for facial recognition. Don't really think it's a 100% accurate and even if it's at 97% accurate, you're going to get some false positives, which is not a great experience. Right. So, and then there's the perception piece. I don't think people really like having cameras on them all the time. Yeah. I think it's a, it's a use case, right? If you go to an airport it's kind of expected. When you go into a commercial office, it's not really expected. So it's about perception. So we're a bit concerned about that. So it was a few things, it was security perception, accuracy that led us down to the hand scanner path and they are highly accurate.
Speed is fantastic and quite simple to integrate at that level. And they also give you the choice of access cards as well. We've got a card reader on them as well. So, we're not saying you can only use your hands. We're saying that you have choices. And I think this is a fundamental, when we start to talk about technology that's customer facing it's about choice.
It's not about dictation to bring them on the journey too. So we integrated the hand scanners into the, into the gates, the gates into the destination control system. That was also integrated into the car park management systems for the boom gates. So the skeet arts system down in the basement, the end of trip, access and locker management system, and also the bike rack storage area.
So as a customer in the building, if you were enrolled on the system, you can literally go from strict tendency without touching anything by just passing your hand under it. Actually, it's a bit you, I mean, I'm happy for you to share the video because it's public, but it's, if you have a look at it carefully, you actually don't need to touch anything.
It looks like you are, but you're not. You're just waving your hand through. So it's a really cool story, but we managed to put it off and the best story is customers love it, you know? Yeah. But the, the challenge was seven technologies, seven different kinds of types of protocols, seven different types of connectivity.
We were lucky that we had, you know, I'm not going to do much of a plug, but Honeywell were the main kind of guys in the building have their security system. They had a pretty good team on site. And so through a lot of collaboration, we managed to get all these systems to talk and share and passed out very quickly.
That's what it was all about. I was very worried about the latency, because if you think about it, when you push your hand, think about the chain effect of hand scanner recognizes you're out in the building, passes it to the lift access control very, very fast.
James Dice: [00:42:49] So what is the architecture of this look like? Uh, if, if I'm picturing like a typical, all these systems are in their own silos, there's multiple levels to system,
what does the architecture look like?
Jon Clarke: [00:43:01] I'll try and remember it from, so from scanner to access control, it's actually acting a bit like a card, it's override B. Okay. But it could also, um, issue, um, the weight and signal. So you can make an app just like it's reading a card, just your hand is your card. From the IP instructor
it goes into the Honeywell API system and the Honeywell API system is pretty much orchestrating between destination control system and they have to write some custom scripts to make it all work pretty quickly. And there was a bit they had to write to integrate the idea management system. So the hand scanning enrollment system into, into platform to make it all come together. So there was quite a lot of coding involved, a bit of testing. We tried that we need to try something else, great project.
James Dice: [00:43:48] So how do you think about then taking that to other buildings? It sounds like it's you know, bespoke a little bit, um, for that one.
Jon Clarke: [00:43:55] 100%, yeah. And one thing I would add is that the most complex piece to get right with the integration of the visitor management.
Okay. We had to create a load of different personas of who's going to come in the building ,and when, what access will then needs to work out how to get that visitor piece to work just as well as if you're enrollment system, because you get to a point about, um, rolling out to the buildings, scalability is a real problem.
We can take the philosophy to other buildings. We could use the hand scanners, but we'll probably have different flavors of technology within those buildings. So therefore we have to start the integration process again. Right. So commercial viability and timelines start to throw a few challenges up, scale it out.
James Dice: [00:44:35] And then in some cases like you wouldn't even have the Honeywell guys under the next building, right?
Jon Clarke: [00:44:40] Yeah, sure. I mean, I think of it this way, even if we did, they'd have different access control or they could have different lifts.
James Dice: [00:44:46] Yeah. Super interesting. Cool. So while we're talking about sort of scale, how are you thinking about as you sort of pilot and sort of converge?
Well, you had mentioned convergence, so why don't we talk about first and then maybe it relates to scale, but I want to hear about how you're kind of thinking about scaling up different use cases. You mentioned convergence of projects, what do you mean by that?
Jon Clarke: [00:45:07] So, um, we carried out a technology audit to work out all the technology across predominantly office portfolio and where it sat in its lifecycle.
And there were a few things that came out of that. And the process that we went through is I created an online questionnaire and there weren't many questions that were, um, free for people to populate. They were pretty much, um, I dictated the answers they could choose from. The reason I did that is because of when we started to digitize the information, I could create a really fast dashboard of what was coming up towards end of life, what had been recently replaced. Okay. Um, so it was a really good starting point for me when I first came into Dexus to understand what am I playing with? Um, but you know, the benefit out of that was that maybe we got an opportunity to have to look at our forecasting.
We can understand when technology is going to come up for replacement. Maybe it's not full replacement, maybe it's partial, but we can start to plan out that maybe a five year, I'm forecasted for what's happening. What I started to say, though, is that there were things like CCTV upgrades that have been planned.
And they were also the BMS upgrades that are being planned and they were maybe a year to 18 months apart. So they were fought in different financial years. And what would traditionally happen is they would be dealt with within those financial years and they'd be almost side projects. So I looked at well,
if I know that the BMS can last another 18 months and it's not a high risk for us, can I move that? And can I then use the CCTV upgrade from analog to IP cameras to create an integrated communications network for the BMS to use? And that's what I mean by converging those projects to get again, economies of scale, get the best out of the combining these projects together. That was one of the strategies that we're
we're trying. I know a couple of projects have gone past and being really successful, worked really, really well. Just, just different, just a different way of thinking about things. I think that's, what's kind of like, I know I've got this word smart in my title, and I think I'm not sure whether I was a 100% comfortable with it, but I think it's not about technology.
I think it's about being smart with your money, smart with, you know, tech, smart with process.
James Dice: [00:47:06] I love that
because people, when people talk about convergence and knocking down silos, a lot of times they're like complaining about the construction process.a little bit, but then you go ahead and do all the replacements with those same silos.
And so that, that's awesome. Very cool. So where do you think we're at? Like, I'm finding myself wanting like some sort of like therapy from you, John right now. Like where are we at of all the things you guys are doing? You know, building new buildings, operating them. You've been at this from many different lenses, right?
Where are we at right now as an industry? And like, what's still holding us back from getting to smart. Your favorite word? That's a loaded question.
Jon Clarke: [00:47:46] I think the industry is in a little bit of state of confusion at the moment, if I'm really honest about it. Okay. Moving so fast, it's a problem for people to keep up with things.
And I took it to you previously about, it's very challenging in a construction environment because of it's locked down to time processes and it could be a five-year build and in five years time, what you bought is now out of date. So I think that there's a real challenge there. And I think the industry is a bit of a waiting game that's going on.
I think if I kind of, you know, related to, to maybe an Australian saying, think about surfing, right? The interface out on the waves, and there's a few catching, some odd waves here or there and they're going forward, but they eventually come back to where the waves are. A lot of people waiting for that big wave to all go in together and all go.
Yeah. And I think that wave is getting closer and closer. I reckon within five years, you'll see a lot more standardization in practices and the ability to deploy tech C more seamlessly than it is at the moment. So I think that we're right at this point in time now where, there's a big thing coming. It's been talked about for a long time and I can actually see it in the distance.
James Dice: [00:48:59] Yeah. So for organizations like you guys, then focusing on that data enablement infrastructure layer is kind of like getting ready to catch the big wave.
Jon Clarke: [00:49:07] Absolutely. Yeah. And future-proofing, you know, as much as you possibly can. I want to, I want to do like a little rapid fire round real quick on a couple of different things.
Uh, before we, before we do this real quick, Where do you sit on ontologies? Right? So you're, you're thinking about data enablement and you're you mentioned standards a couple of times, where do you, how do you guys approach ontologies at the Xs? It's quite new, to be honest with you for us. Yeah, but when you have one at a corporate level, so we've ran out the program over the last few years to digitize our corporate operations cost using a version of an ontology.
So we need one full stop. Right? You've got to have one. If you don't have one, you won't get the power out of all the information. Yeah, I'm not sure about how the industry is going to adopt which ontology and whether they all serve a purpose for different things. So whether they be different flavors with different outcomes.
Yeah. I don't know whether we've gone backwards a bit where we are now in the backnet versus modern kind of world of, is it haystack, is it brick or is it both, or is it a combination? You know, I think it's a watch this space time at the moment. Ontology. Absolutely. They're opinionated. And you just have to have your opinion of what works for you.
I think at the moment, and you know, I was doing a, an industry session yesterday where, um, one of the guys from Google was a keynote and that's pretty much what his message. Anyway, he was saying that, you know, um, just have one, I have one book for the books for you. I actually was thinking that was the main message.
So we will have one that works for us. Absolutely. Okay. Cool. Second rapid fire question. Why don't you feel about, so you talked about analytics, you know, one way of pulling the data to the enterprise. How do you feel about two way? How do you feel about supervisory control and the new types of startups you guys probably get approached by all the time?
Yes. Is my answer. It has to happen. Okay. Why are we ready for it? Yeah, there is out. I don't think that the, again, it's the data structure and it's the learning. I don't think, I think there's too much human interaction required to make decisions at the moment. Totally. But it's definitely coming. let's close this with two truths and a lie.
And I told you before we hit record that I have 100% record so far in case anyone cares. No one's noticed or going to put that to the test.
So, um, my 18th birthday was pretty interesting. I was the only one that wasn't at my party. I used to teach windsurfing, but I can't. these are great. All right, cool. And I did a wing walk for charity. What's a wing wall. What does that you stand on the, on the, on the wings of a pipeline. It's not gonna frame you stand on a frame.
You built them obviously, and then you fly around for 15 minutes. I think that one's a lie. A hundred percent. Those are good ones. Tell me about this 18th birthday party. So I started my career in petrochem and my boss came to me one occasion and said to me, I've got some good news and some bad news. And I said, what's the good news first.
He said, I want to send you to the Caribbean to do a field trip. And I said, what's the bad news. You said Sunday, birthday unbeknown to me, my friends would organize my eighteens. Um, he didn't know that I'd gone off to the Caribbean. They're like calling you like come over for pizza or whatever, and then never answered.
That's great. Very well prepared. I love those three prompts. So thank you for that. And thank you for coming on the show. I really appreciate it. That's good. It's been great. All right. So yeah, it's been fun.
James Dice: [00:52:43] All right, friends, thanks for listening to this episode of the nexus podcast for more episodes like this, and to get the weekly nexus newsletter, which by the way, readers have said is the best way to stay up to date on the future of the smart building industry. Please sit. At nexus labs.online, you can find the show notes for this conversation there as well.
Have a great day.