Turning Construction Data into Digital Gold

Robert Hoffman:

All right. Hello everyone, happy Wednesday, or wherever we are today. Welcome everyone to this webinar, which is Turning Construction Data Into Digital Gold. You’re probably wondering who I am, you’re probably wondering what that title means, if you haven’t read the abstract. But thank you very much for giving us your lunchtime, or potentially your evening if you are joining us from abroad.

So my name’s Robert Hoffman, I’m a director in BDO’s Project & Infrastructure Advisory team, so I work a lot in the planning phase of business cases and projects. Now, thank you very much for those who have joined us from PMI, from InEight, from any organization out there. This is one of the first of many co-partnership webinars and events that PMI Queensland and InEight are looking forward to having together, now into the future.

Now, a little bit of a really quick housekeeping. Please feel free to use the chat box. Let us know you’re here. Let us know if you’re enjoying it or hating it, and you can use this feature chat with each other, but also chat with us throughout the program today. I believe it is, they are checked before they come through, so if there’s any that are a bit curly, not everyone may see them, but please feel free to send them through.

You’ll see there’s a Q&A function as well, in here. Please feel free to place a question into the Q&A function at any time. We’ll try to address as many of your questions as we can throughout the session, but also hold them to the end of the webinar as well, so we’ll try to cover as many of those as we can today.

You’ve also got reaction features, so we do want to see, you can use the emoji function on there if you dare, to react to any of the elements that are being discussed today as well. And please at the end, please give us feedback, in terms of today’s webinar. We really do appreciate your feedback, so there will be a short survey as well, if you have time to do so.

Now, rather than trying to introduce each of these panelists myself, as they are much more incredible than I am, I’m going to let each of them introduce themselves for you. So Matt, I want to start with yourself.

Matthew Macaras:

Great, thanks Robert. Hi everyone, my name is Matt Macaras. I’m the Senior Director of Solution Engineering for InEight in Asia Pacific. My background is through industrial construction, we can say I kind of grew up starting out in estimating, field engineering, superintendent field management, through to project management. And then, seven years ago I joined InEight as part of the technical solution engineering team and now I manage that team in the region, where we help map our customers’ and clients’ business objectives, pain points, and desired outcomes to the InEight solution so that they’re able to achieve that within their business.

Robert Hoffman:  

Awesome, thank you very much, Matt. Thank you for joining us today. Michael?

Michael Maslen:            

Yeah, thanks Robert. Yeah, so my name’s Michael Maslen, a little bit about me. So I’m a major program digital leader at Jacobs. I’ve been in and around capital projects in some way, shape or form since the, gasp, the late ’90s, from various different perspectives. In my current role, I work with clients in Azure, and ANZ, to drive digital innovation and process standardization, aligning technology with business needs. And I lead a growing team of technology solutions specialists across program controls, info management, reporting, analytics and AI. And really pleased to be part of the panel today. Thanks, Rob.

Robert Hoffman:

Yeah, awesome. Thank you very much, Michael. And Graeme, I’m assuming you have something to do with trains and tunnels.

Graeme Newton:

Thanks. Yeah, thanks Rob. Look, I guess I’m coming at it from a project delivery side of things. I’m the CEO of Cross River Rail Delivery Authority, and I’ve been in that role for a bit over eight years now, since the project’s conception. So for those people who aren’t familiar with Cross River Rail in Queensland, it’s similar to the Melbourne Metro, Sydney Metro, Cross Rail in London. Basically using a tunnel to duck under the city, and under the river, and to provide more efficient rail systems. We’ve got four underground stations, a whole bunch of stabling yards, new station upgrades, disability access, those sorts of things. I guess my relevance for this is, I’ve been a big advocate for digital innovation in our project, and we’ve been using a 3D digital model throughout the whole life of the project, but I’m sure we’ll talk about that a little bit later on.

Robert Hoffman:  

Awesome, thank you very much, Graeme. And based on the attendees we’re getting through, already now, I think we have enough to fill a train at this rate. So thank you for those who have just joined us recently, as well. All right, we’ve only got about 55 minutes now, so we’re going to get straight into some of the questions for you all. So this is a question for all of you, because I think this is where we really want to hone in on the topic that we’re covering, which is, Turning Construction Data Into Digital Gold. What does that actually mean?

So Michael, why… Sorry, Matt, apologies. Matt, I might start with you. Starting with the big picture, when we talk about digital gold in construction, what does this mean for you?

Matthew Macaras:

Yeah, thanks Robert. I guess from a digital gold standpoint, really for me when I was in the field and now as I help our customers, that is having the right information at the right time, to make sure that we can affect positive outcomes on our projects. Whether it’s a safe delivery of a project, an on-time delivery, on-budget delivery, preferably all of the above and much more. That’s really what digital gold is, is having that information to be able to have those positive outcomes.

Robert Hoffman:

Yeah, perfect. What about yourself, Michael?

Michael Maslen:

Well, I guess from a Jacobs perspective, digital gold in construction, in a similar way to what Matt was just talking about then, it means taking all the raw everyday data that our projects and programs are spitting out. Things like sensor readings, schedules, field logs, RFIs, anything and everything, and turning that into reusable living assets like dashboards, and digital twins, and the like. Because those are the things that drive real decisions, they drive value, they drive savings on our capital programs.

I’m going to coin a new phrase today, digital dust, and there’s nothing worse than that. And to me, that’s information, or it’s not even information, it’s data buried in spreadsheets and on sticky notes. So being able to harvest information in systems for real value. That’s the gold.

Robert Hoffman:

Yeah, nice. I thought digital dust was when you come back from leave, you open up your laptop after Christmas, and you’ve got digital dust all over it. But that’s a new term that I’ll happily adopt. Graeme, what about yourself? Digital gold in construction, what is it about?

Graeme Newton:

Construction for us, it literally is gold. I mean, anything that we can use the model for that can reduce time or reduce cost on the program. Notoriously, these projects, it’s not so much the physical build but the assurance, testing, commissioning, that’s what takes a lot of time, and what takes a lot of convincing of those particular people who’ve got to have the acceptance. And the sooner you can get into that, the better. I mean, there’s the obvious stuff that’s well known, which is using it for stakeholder engagement, and those sorts of things, public engagement. But for us, really, we’re at that stage where we’re using the 3D model to try and truncate how long it takes for the acceptance and assurance, or assurance and acceptance phase, which then sets us up for insights that can happen prior to the completion of the job. So we’re using the 3D model to allow those people who are going to be doing the assessment, and the acceptance, allowing them to actually get in and tool around on it while the project is actually still under construction.

And the other aspect is, things like a rail tunnel that’s underground, very difficult to get to. You’ve got to get the right certification to get in there, and then if they’re doing testing, you can’t go in there because it’s not safe to have trains and people interacting in that environment. So you can use the model to do those sorts of things. Now ultimately, the final assessment needs to be done in real life, but by using the model, we can actually truncate that process. And we’ve got a model or an approach that we’re taking which allows a cost benefit analysis to be done by our team around ideas. And I can go into more detail as we progress.

Robert Hoffman:

By digital gold, you could mean currency there as well, when you’re talking CBAs. That’s actual dollars.

Graeme Newton:

Absolutely.

Robert Hoffman:  

… as well.

Graeme Newton: 

Yeah. And time is the biggest thing. If you can carve time out, and you get concurrent activities going, that’s the most important part of it.

Robert Hoffman:

And it sounds like you’ve obviously drawn insights from what you’re doing at Cross River Rail, and using that to your advantage. Matt, probably a question for you, given you’re from InEight, and I think many organizations do say they’re data rich, but insight poor. That’s obviously not Cross River Rail, which is almost the flip of that. What’s the first step for flipping this equation in your opinion, Matt?

Matthew Macaras:

I think the first step is really identifying what you’re trying to achieve. What are the desired outcomes from a business perspective? What are the business objectives that you’re trying to get to? Whether that is, as we mentioned before, on-time delivery of the project, and on-budget delivery of the project. Whatever those key objectives are, identifying what they are, and then looking at what the outputs are required to achieve those objectives. So if we need instant visibility into productivity data, or productivity information, so we can make real-time decisions to affect what’s happening in the field. Or if it’s instant visibility into our contract packages and what’s happening there, so that we can then make the appropriate procurement decisions.

Identifying what outputs feed and will ultimately get you to realize those desired outcomes is the first step in that process. And then from there, you can identify and look at what are the processes that feed those outputs, and the data that’s required to feed those outputs. And then anything else on top of that is really data for data’s sake, and it’s not helping you achieve your outcomes, and it’s becoming a burden on the project or a burden on the organization.

Robert Hoffman: 

That’s a good point. I think you’re really going to the areas of, we’re getting into sort of a live project environment here as well, aren’t we? And I guess Michael, from your perspective, when you hear “digital gold,” what does it actually look like from a live project point of view?

Michael Maslen:  

Well, as I was saying before, that digital gold from our perspective, on a live project or program is, yeah, taking those raw daily data feeds. Was it sensors or progress logs, or schedules or time sheets, or contractor activity schedules, feeding that into live dashboards. Because it’s not a nice-to-have anymore. Every program needs that instant team-wide decision-making ability, it’s not some ethereal abstract idea, this is just how we work today. I mean, having this information on tap is just so critical in today’s major infrastructure environment, because these mega-projects that we work on are just too complex and risky for gut-feel alone. A bushfire or a delay or a storm can drop millions. So being able to harness that real-time data if you will, for predictions and alerts, can make or break program delivery or on-time, on-budget delivery. So, yeah. It’s a very real, very immediate, like I said, not an ethereal concept at all.

Matthew Macaras:

And I think Michael, I was listening to a keynote actually earlier today, where it was being talked about. As you said, having those data points available to you, 20, 30 years ago, we would get cost reports on a monthly basis and that would give us our snapshot of how we did for the month. But as projects are becoming more complex, as they’re becoming larger, we need to be able to see the status of those projects on not just a daily, bi-weekly and a daily basis, but sometimes if we’re talking about a shutdown on a major piece of infrastructure, we need to know down to the hour or down to the minute what’s going on.

Michael Maslen:  

Exactly. And on water jobs, if we can’t pull SCADA data and GIS data from the plants into resilient views, it’s impossible to keep services running.

Graeme Newton: 

I think it’s a good point you guys are making, because the gathering of the information, I’m at the build phase and the install phase, but what you’re talking about there is the operational phase.

Michael Maslen:  

Right.

Graeme Newton: 

But without gathering the data, the core data, the foundation data, at that early stage, often the data build side of things is an enormous task, and that becomes cost prohibitive. And a lot of organizations balk at that. Most people I speak to who are in this field, your peers, the biggest barrier they’ve got is convincing their bosses, or their clients, that they need to actually spend the money to gather the data into a form that it can actually be interpreted. And that seems to be the constant frustration across the board. Whereas, if it can be gathered in a form and in a common data environment, we’ve been gathering data for over seven years. We didn’t know what the data was going to be used for. We knew we had some initial use cases, but we didn’t know what it was going to be used for. And I can’t tell you what it’ll be used for in 5, 10, 15 years time. I hope it will be used for those things that you’re talking about, the SCADA, the real-time. It may well be maintenance, monitoring or whatever.

But we’ve been able to capture pretty much every element of this project, because it has to be digitized, we’re capturing it in that common data environment. It’s a repository that’s owned by the state, the state government, and so future owners can then utilize, or future operators can then utilize that data in that environment. So it’s kind of one of those things, gathering it at the source is good, knowing that contractors and their designers are going to be gathering that anyway, so formulating a contract environment where that data is actually owned by the client, not by those consultants and contractors. And it ends up, as you said, digital dust in the bottom drawer of the designer’s PC, because they’d have no further use for it once they’ve finished that job. But for the owner it goes on forever, and a one and a zero can be used in many, many different forms, as long as it’s able to be tapped into.

Robert Hoffman:  

Yeah, it feels like Graeme, you’re an exemplary case study at the moment for what you’ve got going on at Cross River Rail, but I can imagine it wasn’t that easy to get to this point, right? And you talked about blockers, barriers, that… How do you convince someone? Is it sort of showing them the insights, then like you said, [inaudible 00:15:19]-

Graeme Newton:

Yeah, look, I think it comes with leadership also, all the way to the top end. I’ve always had a bit of an affinity towards this type of thing, but also we had a meeting with some people from Cross Rail in London, and they got into this gathering data. But they started late, and they said it was really hard to capture it. So it’s kind of like, you go away and leave and come back, and you’ve got to clean out your inbox. I mean, the sooner you start and the sooner you curate it, the easier it is going forward. And we set up, as part of our contract set up an arrangement where regardless of what the data is, it gets put into that common data environment. We didn’t know and don’t know what it would be used for, but we’re now using it for everything from train derailment recovery models. Or we used it for the signal committee for Queensland Rail, to look at what the ETCS sight boards would look like as well, too.

So it really varies, and you just don’t actually know. We used it for some Indigenous work where we were dealing with First Nations people, looking at what the terrain was like before and in current environment. So you can use it for so many different things. But at the moment we are absolutely sleeves up focusing on the pre-testing and commissioning, and then the acceptance and assurance, and you have to run through all sorts of different scenarios. Things like train on fire and the tunnel, how do people escape? Which way do the fans blow? Those sorts of things. We can run all of those scenarios with the data that we’ve got available to us.

Robert Hoffman:  

I think I’d be a bit, I wouldn’t be doing my job properly if I didn’t ask Matt a question now, particularly given being from InEight, and having a platform that could potentially solve these problems. And I must admit, the complexity of my platforms or tools are sort of in the Office 365 world. So Matt, keen to hear from yourself, from a technology perspective, what role do integrated platforms like InEight and others play in bringing what Graeme’s talking about together? Helping to turn raw data into-

Matthew Macaras: 

Yeah, and Graeme’s brought some really good, real live cases, and mentioned a few just there. I think some of the roles that they play that are even back a little bit further, before we even start, able to run the scenarios that Graeme’s talking about, is being able to put the information in once into the system, and then be able to reuse that multiple times. Graeme’s just talked about, “We don’t even know what we’re going to do with all of the data that we’re capturing yet, but we do know we’re going to need that at some point. So if we get it into that common data environment, if we get it into our integrated platform, we can then reuse that information multiple times, and not have to capture or input that into System A, System B, System C, which spends a lot of extra time, a lot of the extra storage requirements. So that’s one of the main use cases for an integrated platform is being able to input once, and then reuse.

It also allows you to connect different sources of data, whether it’s schedule and cost, or it’s documents and the 3D environment, or connect all of those into a single common data environment. And what that does is, it lets you minimize the gaps, or minimize any blind spots that you may have on your projects or in asset operations. So examples of that would be like contract leakage, where we’ve approved a change order for a subcontractor, but that actually affects our overall scope for the entire project, and we’ve just forgotten to add that to the next change order, or to the next budget round when we’re trying to go get some more budget to finance the next stage of the project. Capturing accurate progress then enables us to have accurate forecasting of where we’re going to be at, and actual real-time visibility as to our current status, so that we can know what’s going to affect things going forward.

So having that connection, minimizing those gaps, getting rid of those blind spots, lets us then manage those things going forward. And integrated platforms, really, that’s what they’re meant to do out of the box. And then that gives you those immediate insights, so that you can take action on those, and not wait for the tabular reports to come out in four or six weeks, or eight weeks, depending on how quick your month-end cycle can be, to actually take effect on those and turn positive outcomes. Or, either optimize situations, or reduce the risk or the negative impact of issues that have popped up.

Robert Hoffman:            

Yeah, understood. And Michael, as a customer of InEight, when did the rubber hit the road for you, and when did you find that data started to become more of an asset than a byproduct of delivery?

Michael Maslen:  

Well, on a major water program that I’ve just come off, that transformation… I’m just going to talk a little bit more broadly about the transformation first, and I’ll come back to your question about when data starts feeling like an asset. But that transformation starts way back with the program team, during the implementation of the tools to support the program having that realization that, “This is actually happening. We are actually doing this.” There was no going back. I mean, what’s the phrase? The burning of the boats, right? Between that, between the user acceptance testing, the go-live, the live parallel run, the reality of what is upon the, in this case, on my program’s example, the controls and the commercial teams.

Now arguably, it had been their first go-live using, in this case a modern PMIS, being InEight, was when that light bulb went on for them. And that transformation was around the, “We’re actually owning, I’m owning this, we’re owning this. This is ours now, we need to run with it.” So as the team settled into that regular pattern of usage into the system, that go-live sort of transition, that manual reporting process that had dropped from weeks to days to minutes, and that’s obviously, there’s obviously big savings and saving of the program dollars and cents, from that perspective. And the InEight platform was absolutely crucial to that data flowing straight into the system.

So when data started feeling like an asset is absolutely all about the reporting and analytics and information, and the decision-making ability at the back end. But that transformation is, and I’ll harp on about this quite a bit is, it’s much about bringing the team with you, because they’re obviously a crucial part of that transformation. Just looking at the data, just specifically and how that transforms into an asset. I mean, the game changer for us was cleansing and blending source data, and in some cases source data that, “Why are we capturing this? Capture more rather than less,” to Graeme’s point. “We don’t know exactly how we might use this, but if you’re capturing everything very early on it as a higher resolution as possible, then you’ll find a use for it. Trust me, there’ll be gold in that CDE.” So capturing that data as early as possible, as fine-grained as possible, and that data and that information starts paying dividends month after month.

Robert Hoffman:            

Yeah, fair point. I mean, talking about digital gold, I feel like we are missing some digital gold in the form of questions from all of those online. I think we’ve almost got a hundred people online, and I’ve only seen a couple of questions come through, so if you do have any questions please feel free to put them through, and we’ll ensure that we do get them answered.

But there has been one, so I don’t know who’s best to address this, but we might start with you, Graeme, given you’re in, I guess the operational phase at the moment. But, “Where does data privacy and sharing fit into all of this?” And I think this might be a key thing for you in your broader open data environment.

Graeme Newton:           

Yeah, look, I mean ours is a contractual arrangement between ourselves and the contractors. So all the data that’s produced is recognized as being owned by the state, so it’s related to all of the assets that are being built, the stuff that’s been captured. I mean the state’s paid for it, we’ve used taxpayers’ money, so therefore there’s a very good reason why it should be the asset of the state. Whenever we share that data with another agency, so it might be another government department or local government, say Brisbane City Council or something like that, we always have a data sharing arrangement with them. An actual, it seems like every other day I’m signing a data sharing agreement with another agency or another group. And that’s really quite essential because it articulates who owns what data, how it can be used, when it can be used, and what permissions are required as part of that.

So I think, don’t miss the contractual side of things. So then, your sort of question in relation to privacy. Well, privacy is sort of covered off in that regard. I mean, we don’t gather personal information about individuals, or anything like that. It’s really more about the asset. If you see it as a, I suppose we all know what a physical asset is. If you see it in the same light and say, “The ownership of it is the State of Queensland, managed through the Cross River Rail Delivery Authority, and anybody who wants to use that data for whatever purpose needs to have an agreement with us. We don’t make it freely available.” But that said, if it’s another government agency, we don’t charge people for that. Because it is, my view is, the taxpayers of Queensland paid for it, so therefore if it’s another government department using it, well if we can save that department spending more taxpayers’ money and use that asset.

So for example, we’ve provided significant data and information to the Olympic Infrastructure Authority, also to the groups that are doing, the Department of State Development who are doing the arena around Woolloongabba. We’ve got a lot of, because we basically go through the middle of the city, so wherever there’s a piece of infrastructure or there’s something that’s related to us, we might have captured it, we share that information with others as well, too.

So I’m more than happy to do that. We work with accessibility reference groups, and we also provide them assistance there. And then we’re also working with other agencies around things like wayfinding, and the interconnectedness between different transport types or different assets, and pedestrian routes and things like that that are accessibility-friendly. So there’s many ways in which we can share it without actually putting it at risk, that it’s going to bleed out into the wrong hands.

Robert Hoffman:            

Yeah, really good point. Matt, Michael, any further comments on that, in terms of data privacy as well?

Matthew Macaras:        

I guess the only thing I would really add is that there’s probably two ways to look at data and data privacy, and there’s certain sets of data that you’re going to want to share and have, and be that single source of truth for everyone working on that particular project, or everyone working on that particular asset. And you do that to make sure that everyone is working on the same page, working towards the same goal, using the same information to minimize rework, and to maximize the delivery of that asset. And then there are certain things, contractors will want to keep their own books private, which is perfectly normal and acceptable. And owners will want to do the same thing. So I think depending on the type of data that you’re using, some of it you’re going to want to share and use as that common data environment for the, we’ll call it the broader project ecosystem of all the contractors, and the owners, and consultants working on it. And then each group will have their own bits and pieces that are naturally going to be private to them, that won’t be shared.

Graeme Newton:           

Yeah, I’d sort of jump in there. Just remember, like I said, if the state’s paying for, it’s a state asset. So therefore, and I get what you’re saying, that some contractors, particularly the hard dollars, might want to hold some numbers back. But for those people who are from state governments, online, I’d be saying the point at which you’re doing the procurement, the tender phase, is when you’ve got the ultimate leverage and they’ll agree to anything. But in an ironic way, people who work in the digital environment, like Matthew and Michael and co, have given us feedback and said, “Thank God you wrote that into the contract, because it meant that our masters were forced to gather that information and make it available, we’re able to use it in our environment as well too.”

So we’ve seen one of our suppliers, through an alliance, has done some pretty significant innovations as part of their… Things like utilities tracking, and things like that. Had they not been forced through the contract to gather the information, there would’ve been some cost-cutting there. I mean, civil contractors notoriously want to try and cut costs, and they usually cut it out of design. So that’s where they’ll try and do it. And if they can avoid spending money on gathering ones and zeros that they’re not going to be able to monetize, then they’ll actually carve it out. So as a client you can force that, but also force the innovation as well, too.

Michael Maslen:            

Look, the only thing I’ll add to the discussion, is that every project runs on digital information. So data privacy obviously matters. We’re holding contractor records, employee records, 3D, 4D, 5D models, security layouts, drone footage. All of this information is naturally sensitive. Obviously, any breaches of security, or cyber security breaches, obviously can trigger reporting and contractual liability as Graeme was talking about, and project delays. It’s no longer… Look, all I’m going to say is, to wrap this point up about data privacy, if you will. It’s no longer just an IT issue. It is a core commercial and risk management priority for major programs, full stop.

Robert Hoffman:            

Appreciate the insights. I feel like we’re about halfway through, so those in Brisbane, you’re probably about halfway through that sandwich you’re having for lunch, and it’s been way too positive so far. So I think we do have to talk about some challenges now. And Michael, while you’ve got the mic, what are some of the biggest challenges you’ve faced in collecting, cleaning, and connecting project data?

Michael Maslen:            

The biggest challenges in this regard that I’ve witnessed, we’ve all faced them, right? It’s silos from fragmented sources. On a major program, the lens that I bring is from typically program controls, or project controls. So it’s about trying to bring in time sheet, or cost, or schedules. Contracts’ change information, risks, issues, et cetera. That historically have been stuck in different tools, and I’ve encountered absolute sheer messes. Is that right? Is that the right grammar? Of dirty and inconsistent data at scale. I’ve seen the best of the best, but I’ve also seen the worst of the worst, on these mega and giga programs. I’ve had people look me straight in the eye on a major program that I’ve just come off, professing that the data and the structures that they had were absolutely pristine. And if you’re in spreadsheet land, I mean, there is that belief still out there amongst our market, and unfortunately our peers. That if it’s in a spreadsheet, it’s controllable. Sorry, it’s not.

The program that I’ve just come off. Very early on, the team was wasting hours manually pulling cost reports and documents from spreadsheets, and risking errors, and the classic challenges with spreadsheets, et cetera, before we put our system input, before we put InEight in, as it happens. It was not just InEight, it was also doing all that cleanup of data, and building a data warehouse and all the ETL work, to cleanse that information before it was coming into the CDE.

Yeah. It is never really smooth sailing, but I think the biggest eye-opener for me is those that don’t observe the problem for what it really is. And sometimes that’s the biggest hurdle is that mental leap to say, “Well, actually, there’s work here. There’s work that needs our sorting out.”

Robert Hoffman:            

Yeah, that’s a good point. And Matt, I guess from a platform point of view, what’s some of the things that you’ve seen in terms of some of the biggest challenges that your clients face?

Matthew Macaras:        

Similar to the things that Michael mentioned, right? It’s having consistency in the data, or finding consistency in the data that they’ve got. User change management is another big one. People grew up using spreadsheets, right? I mean, I’ve got a six, no, he’s an eight-year-old, sorry. I’m getting in trouble for that one. I’ve got an eight-year-old son who, he’s learning how to use PowerPoint and Word and Excel. So it is ingrained in us from very early ages that how to use those systems, and everyone’s comfortable with those. And then taking that experience, and putting it into a new platform that provides the outcomes that the business is looking for, but also provides the governance around it, so they can’t just go in and change a number in Excel willy-nilly, is confronting to some people and requires some change management, and some explanation of why the organization has gone down that route.

So yeah, having that consistent data and then managing that change across the teams that are adopting the solutions, is important.

Robert Hoffman:            

Graeme, I have to ask you then, I think we’ve covered some of the challenges. It does seem to come back to, I think consistency was a pretty common theme there. Discipline of people, in terms of how they interact with whatever the platform or tool may be. How do you find that, from your perspective, is the greatest challenge from the technology itself? Or is it more from the getting people to be consistent and disciplined in how they use it?

Graeme Newton:           

Look, I wouldn’t pretend to know the intricacies of how it gets used. All I know is that the use case gets asked for, they turn up with it, and they do, the magic happens somewhere in between. So I take my hat off to these guys who can do it. Look, I think making sure that, as these guys are talking about, the information is accurately captured in the first place, and that it’s managed effectively. The technology, I mean, remember. We started this seven years ago. If I showed you, side by side, what some of the visualization looked like back seven years ago, and then how some of the buildups could have been done then, versus what it looks like now. I mean, we’re basically at photo real stage now, whereas back then it still looked a little bit cartoonish, still the same ones and zeros, but basically it’s been groomed and refined, and more applicable. And you can get in the granularity, the specifics are in there.

And then it comes down to, “Well, how do you then turn it into something?” I think these guys, Michael was talking about using it for say, program control, or reporting and data. There is a key, we are gathering data on progress, and it does require, I mean, something needs to be installed, and the piece of data needs to be installed into the common data environment. And if there’s a lag between that, you are getting missed reads on that, on a particular item that might’ve been bolted down. And then it gets tested, it takes that person who’s doing that work to actually do it. And so how you can close the gap between the install, and the test, and the results, and getting that into the environment, that’s a key component.

Because we do find there is a little bit of a lag between that. And if you get, ideally you want people with an iPad who are doing it on the spot as it happens, but in reality, you’ve got a tradesman who’s a sparky or a plumber or whatever who’s installing something, they’re too busy focusing on installing what needs to be done. Then it requires someone else to gather it. And then you get into the system checking, really the more intricate components of it might be, tunnel ventilation systems, extraction fans, and then getting into the SCADA side of things. That’s when it becomes more digital, but again, it still requires the data to be installed, captured, placed into the environment, and then be able to be used.

So yeah, I think we’re on an evolution. I think the technology’s really coming along a long way, but I don’t think that the technology is actually the limit. It’s actually the gathering of the information, and making sure it’s accurately captured, accurately put in a place that can be gathered, and it’s done in a timely way that it’s actually useful.

Robert Hoffman:            

Yeah. I feel like that’s almost too-

Matthew Macaras:        

And Graeme, I kind of think about that as two steps, right? Consistency and discipline, which is what you’re talking about, about the people actually doing the work and then capturing the information. Having the discipline, the consistency, to do that. But then from a technology standpoint, also making sure that we’ve got that standardization in there, so that we don’t have someone saying they’ve installed one each of something, and it’s spelled E-I-C-H, and then someone else has spelled it E-A, right? That standardization that you build in allows all the reporting that Michael’s talking about, and then also helps with that consistency in the discipline, because people aren’t having to make complex, or even fairly simple decisions, about selecting between two different eaches in the drop-down list. So yeah, consistency and discipline of the people doing it, and standardizing on what you’re capturing and how you’re capturing it, makes that consistency and discipline that much easier.

Michael Maslen:            

Yeah, because without ease of use, you get that frictionless, the gap between the activity concluding and the data being captured, becomes less and less. And less. And there are organizations that are out there doing it well in our sector. I’ve had the benefit of working for them, and have worked for them in the past. And to me, a large portion of that is, are organizations willing to drive that level of change, that level of discipline, and ingrain that into the teams whatever those might be, to work efficiently. Because the benefits of being able to see on your handheld what your crew’s productivity is, for instance, as the work is being concluded is immensely powerful. But to get there, there’s an uplift to get there. No doubt, no doubt.

Graeme Newton:           

Look, and it’s pretty relevant to a conversation that’s sort of going on in other spheres around productivity improvements in the construction sector, and it’s something that’s been around for a long time. I know Infrastructure Partners Australia are talking about that as well, there’s a productivity commission in Queensland around that. And I guess a lot of people will shake their heads at the construction sector saying, “We’re still using old methodologies, still tying, still fixing and pouring concrete where they could be doing a whole bunch of other things.” But really, I think the digital side of things is where you can actually make some leaps and bounds, where you can bring efficiencies in and insights. And then what that’ll do is start to lead to innovations and going, “Well, do you know what? Using digital modeling, we could actually look at how we could do something differently.”

Robert Hoffman:            

Yeah, interesting. I feel like we’re almost talking around some common misconceptions here as well, and I think, Matt, you might have pointed out that it might just be easy to blame the tool, blame the platform. And then there’s obviously the public assumptions around where some of these limitations are, but are there any other misconceptions that you have all seen around digital transformation in construction? Anything that’s really obvious, that people should know?

Michael Maslen:            

I’ll just jump in. Are you going to jump in there quick, Matt? Sorry. For me, one of the big misconceptions about digital transformation in construction is the need to rip out all of the legacy tools and start, buy shiny new ones. Right? I’m the first person to want to buy the shiny tool, but sometimes realism kicks in, and it’s not always true. You’ve actually got to review and evaluate and integrate, in some cases, what you have. And in some cases that might actually mean the pragmatic approach is to take whatever legacy data is out there, and pull it into a central hub without disruption. At least initially, potentially. Like on an airport job that we worked on, we took [inaudible 00:41:48] from a dock management system, we took schedules from a scheduling system, we took other artifacts from SharePoint and we pulled it into a dashboard. So no full replacement, but just smarter ways of integrating that data, and remediated those systems over time.

I think the last point I think I just want to make really quickly, is that technology alone doesn’t fix everything. I mean, I’m a technologist first and foremost, but over the last 10, 15 years, I’ve come to the glaring and obvious realization that data quality and human buy-in matters more. That, to me, is really where the uplift is, is bringing people on the journey. The technology works, all technology works. It’s getting the people to have the organizational and personal will to adapt to those changes.

Matthew Macaras:        

And that’s the exact point I was going to make, Michael. Is it’s not necessarily, technology is not going to be the end all, be all to solve all your problems. It will help solve those problems, but if the organization and the people are not ready to embrace that, maybe embrace new ways of working, or even just standardize and do the work that needs to be done to make the technology work, then the technology is not going to work for them. It’s not that the technology doesn’t work, but the people and the organization need to be ready to adopt and adapt to get the most out of it.

Graeme Newton:           

Yeah, I think the one thing I’ve tried to avoid and as I mentioned before, is how I let the technologists know that they’ve got the support, we focus on what it is that we’re actually needing, to make it applicable. But how they go about it, and how they gather it and so forth, is really something. You’ve got to let them innovate, come up with their ways in which to do it, or they’ll tell you, “No, it can’t be done,” or it can only be done in a certain way. But be really clear about what’s the end use, what are we needing there? And then what we’ve found is that’s then stretched people’s minds around how to use the technology, and they’ve gone and found other pieces of technology, or evolutions, or spotted other things that could be applied in our environment.

And we’re not in a constant stage of build. We’ve moved through a lot of the construction, we’re moving into the testing and commissioning phase. So the need, just as you’ve sort of mastered the capturing of the build stuff, you no longer need it and you move on to the next phase as well, too. And that’s why I sort of talk, how do we allow people to get this insight early, so they can be gathering this data?

Robert Hoffman:            

Now, I don’t know about all of you, but I’m pretty pumped for the next little while, so I know 2032 is coming here in Brisbane. There’s a lot happening, Cross River Rail is not far away, as well. There’s a lot coming up, and particularly from an infrastructure point of view. But from a data and AI project delivery point of view, what do you see as the biggest opportunities using data, AI more broadly, in the next three to five years and beyond? Maybe start with you, Matt.

Matthew Macaras:        

Yep. I think the first thing to think about when we’re looking at AI in data, is that for AI to really provide the most benefit, or the benefits that people are looking to get out of it, you need to have good data. It needs to be clean, it needs to be trusted. The better your data, the better outcomes you are going to get from, whether you’re using AI or just dashboards and reporting like we do today, or machine learning. The quality of that data is going to be paramount to getting good outputs.

And also, that external data is not going to tell you about your own business. It needs to be good, clean data from within your own business. Everyone does things… I mean, if I want to go install 2 1/2″ to 6″ carbon steel on a power project, yes, there are general ways that that is done. But every business captures that information, the amount of work, the cost, all of that a little bit differently. So you can’t just go grab open source data and say, “Yep, this is going to give me exactly the outputs that I’m looking for.” You need to rely on some of your own, and preferably all or most of your own data, that is good.

But once we’ve got that, I think we can then leverage AI to give us some, what I like to call practical AI. Actual insights, things that are, trends that it is noticing, notifications. Use it to enforce governance, and apply governance across projects. Being able to ask it, “Hey, I’ve got a potential change that a contractor, a subcontractor has submitted to me. How does this fit into my executed contract document?” And have it actually step me through, “That looks like it’s in scope, that doesn’t look like it’s in scope.” Or, “This is the process to facilitate that change order.”

And also using AI to help automate things. Moving manual processes to automating deliverables, but using… Automation is not going from nothing to the final output. It is getting us, kind of like what we use, we can use Copilot for in Office and Outlook today. It’s a great way to get that initial draft, using information that you’ve already got in the system, or using information that already exists in your organization. But it gets you to that draft stage, we still need to apply that human intelligence on top of that, with all of our knowledge, with all of our experience, to make sure that that is applicable for the scenario that we’re in now.

And I think the other one that it’s really going to help out with is project health, early warnings, and giving us some of those predictive insights so that we can make some decisions and take action on things to ward off oncoming risks and oncoming issues, based on what we’ve seen in the past.

Robert Hoffman:            

Michael, what are your thoughts?

Michael Maslen:            

My perspective is that a well-implemented and modern PMIS program management information system, a modern PMIS data and the information that it stores, is fuel for AI. Matt, I agree with your comment, right? Having that clean, metadata rich, time-stamped information is an essential. Twenty, 30, 40 years ago, it was garbage in, garbage out. It’s the same thing, right? It’s the same thing. It’s just how we get there.

From a Jacobs perspective, the way we see things is, the biggest opportunities for data in AI, particularly in program or project delivery over the next three to five years, is in and around autonomous AI agents, predictive digital twins that harvest that raw data into self-running foresight. And none of that’s possible without having the data as a starting point, as I just mentioned, having that clean timestamped data at the start. The practical application of that is, and we’re already building this ourselves and seeing our clients going down this path, is having agents, AI agents, auto-handling RFIs and making tweaks to schedules, and flagging risks, and forecasting disruptions before they hit.

We’ve invested heavily in AI, as you would expect, and we’ve built tools that evaluate and scan contracts and plans in minutes. Flagging issues, cutting review times, and the like. So that three to five year horizon is happening right now. Whatever we’re thinking about doing, out there, someone’s already doing it. But the key is, to build on Matt’s point, the key is have a clean data set. It’s a precursor to everything.

Robert Hoffman:            

And, Graeme.

Graeme Newton:           

Look, I think there’s a whole range of areas where it can fit in. I mean, right back at the commercial phase, getting alignment around what you’re wanting to achieve with your contract to what is being negotiated, and what’s being put on the table. Getting help in that sort of space where contractors will want to squeeze in different caveats, and conversely, the other way around. Owners will want to put in particular things, so contractors make sure that they’re not missing something and under price.

Similarly though, when you get into it continuing through with where we are at the moment, I mean there’s tens of thousands if not hundreds of thousands of pieces of content that need to be checked and validated as part of the assurance process, and that’s just huge volumes of material that needs to be validated. And using AI to help generate that in a form that can be consumed by the regulators, or by the people who are going to be doing the assessment, but equally helping them do their assessment because it all takes time to get through that. And that’s where we have big delays at the end of these projects, where you’re waiting for your assurance and your acceptance, and you’re trying to get through that.

In a practical, physical sense, I mean, we’re already trialing where you take a 3D camera into the site itself and scan, and it basically detects what’s changed since the last time you were in there, and highlights things. But also, I could see where you’ve got non-compliance elements, where a certain step height might be the wrong step height, or it might be a certain width is the wrong one, and it’s giving you alerts that the as-built is not as per the specification. And so, it gives you points to sit down and get it resolved, or overcome non-compliance elements as well. Or also track, like we were saying before, something may not be captured because the person capturing the install hasn’t installed or provided the data correctly, whether it’s with the wrong acronym or the wrong title, or something like that. You’re able to pick it up and go, “This anomaly is…”

So I’m pretty optimistic about how that’s going to play out in the sector, and like I said, everything from commercial to assurance, but also to the physical build, there’s a lot of opportunities there where you can use your digital twin to compare to what’s the physical twin, and then how’s that all sort of playing out? And what needs to be done, what sort of rectification, and that speed to the breakdown. So you’re not relying as heavily on a small group of eagle-eyed individuals who are walking around with a clipboard and trying to find defects, and then having disputes. It becomes far more clear about, “Well, this is a non-compliance, why is that the case? What’s the basis for it?” And it speeds up the discussion.

Robert Hoffman:            

A hundred percent. For three very passionate people, it’s going to be very hard to get you to do, but really quick response to this question. One practical piece of advice you’d give to someone starting on their digital journey. I might start with you, Matt.

Matthew Macaras:        

Yeah, I’ll kind of split it into two. First, identify the problem that you’re trying to fix. And then, start now. It’s the best time to start, the longer you wait, the longer you’re going to wait until you start seeing the benefits of it.

Robert Hoffman:            

Perfect. Michael?

Michael Maslen:            

Practical advice. Well, embarking on any sort of innovation journey isn’t a solo pursuit, right? It’s a team sport, it’s a collective odyssey, as I’ve been known to call it. You have to take your team, your stakeholders, and skeptics along for the ride. That’s vital. You’ve got to communicate with clarity. You’ve got to share the why at every point, you’ve got to over-communicate, whether it’s workshops or updates, or water cooler chats. The destination, we all want to get to the destination, but it’s only when everyone’s on board is when the real magic happens. So it’s not a solo pursuit, it’s a team sport. Get on board.

Robert Hoffman:            

Thank you. Graeme.

Graeme Newton:           

Yeah, look, I’d come at it from a slightly different angle which is, your biggest challenge will be overcoming the perception of it being a cost center. You will always be seen as costing the project, or the business or whatever, and eroding the overhead. And so you really, at the start, you need to advocate that it needs to be gathered right from the start. So there’s not a cost to gathering the data, it’s actually happening incrementally, but equally build up the use cases and the justification how they’re going to actually save time and save money. And that’s what will get both your owner and your project managers bought in, because that’s all they focus on is, “How much is it going to cost, and how much time am I going to save? And does that time that you’re saving me, going to translate into bottom line gold,” that we talked about at the very start.

Every person I talk about in this space, talking about digital twins, or data or digital, their biggest headache is convincing people that this is a good idea. And I see it in the government world all the time. These projects become stranded, because the guys who are pitching it, guys and girls who are pitching it say, “Well, if you give me a couple of million dollars, then we’ll be able to gather all the digital requirements for all the assets around, and then these are the wonderful things we can do.” And the problem they’ve got is they’re trying to make two sales. They’re trying to convince them that that sale in the first place, going through that process and gathering the data, is a good idea. Because the second sale is the one where they’re going to actually make the money, and if somebody’s not bought into it and they’re not an advocate for it, they’ll just go, “A couple of million dollars? I mean, I could build a school for that, or I could do something else.”

So that I think is going to be the biggest challenge for people in the sector, is to build those use cases, understand, speak the language of the person who’s going to be the beneficiary of it. Don’t get hung up on the technology. Like I said, with my people, I don’t really care how they do it. All I want to know is, “This is the use case that I want, and this is how it’s going to save me some time and some cost. Go away and do it, and demonstrate to me that you can do it more cost-effectively than if it went down a traditional path. That’s all I care about. I don’t really care which computer you’re using, and,” nothing against you, Matt, “whether they use InEight, or they use some other platform.” We don’t really care. That’s the reality of it. But if they come back with an outcome, off you go, we’ve got a value for money proposition, and that’s what I’m interested in.

Robert Hoffman:            

Yeah. Awesome. No, such good insights. And Matt, Michael, Graeme, thank you very much for your time today, for sharing with me, for sharing with the entire audiences with us here. We really, really appreciate your time. And yeah, such a good wrap-up. I just can’t get the term digital dust out of my brain now, so thank you for that. I’ll take that for life. And thank you also to those who dialed in today and joined us during your lunch hour, joined us in your evening, whatever time it is wherever you are. We wouldn’t have been able to have this event without yourself, and also the broader InEight PMI team, for helping bring this all together. From Rose and Kenny behind the scenes, huge, huge thank you.

Now, some of the questions we didn’t get to in the chat, apologies, we will respond to those offline. So rest assured we will come back to you, but obviously, we had too many exciting things to cover in this short period of time.

For those who are interested in hearing more, please feel free to jump onto the InEight website for more content, you can see their new technologies, see their upcoming webinars as well. Same with the PMI Queensland website, and socials as well, you can see what’s coming up in store with PMI Queensland also.

Now, please feel free to leave any feedback via our survey, which will pop up on your screen shortly. Thank you again, Matt, Michael, Graeme, and everyone who joined us here today. We really appreciate your insights and time. And until the next webinar, we’ll see you soon. Thank you all.

Graeme Newton:           

See you.

Michael Maslen:            

Thanks, Robert. Thanks, all.