Journey to Industrialized Construction

[00:00:00] Roberto J. Arbulu: So, without further ado, let me pass it over to James, who was initially or earlier introduced. James is going to moderate a session about the journey to industrialized construction. So James, over

[00:00:17] H.J. James Choo, PhD: to you. Thank you, Roberto. So what we’re going to do, what we have prepared here is a group of experts that are going to talk about their specific journey and their role that they play in the journey to industrialized construction.

[00:00:34] H.J. James Choo, PhD: But before I start, I might have to do some push ups to match Amy’s intensity and energy level. So I might have to do that over a bit. So how does industrialization fit within the modern construction discussion that we’ve been having today? And Todd actually presented this earlier. So modern construction has three pillars, which is automation and industrialization and digitalization.

[00:00:58] H.J. James Choo, PhD: And we’re proposing that the framework underneath it is operation science that we, a lot of the speakers, actually have touched on today. So, one of the things that people started to talk about is why we need to actually modernize construction and what is the, and as you can see from the news articles everywhere, that there is no shortage of people diving into it.

[00:01:25] H.J. James Choo, PhD: So someone who actually has the construction user’s rhyme table has predicted that the modular market is to grow to a $215 billion market by 2020. That’s in four years, well, almost three years now. Okay. At the same time, there are large investments being made into the industry, so Parkshire Hathaway is already backing up a startup to partner with architecture of modularizing hotels and apartment buildings.

[00:01:57] H.J. James Choo, PhD: However, the road to success doesn’t seem to be without its own challenges, as you might very well know. A company that had invested a tremendous amount of money has recently filed for bankruptcy. And as you can probably say, this is not due to a lack of funding. Okay? So one of the things that we need to figure out is what is, what do we need to do to make sure our journey is successful to achieving industrial life construction and modernized construction?

[00:02:35] H.J. James Choo, PhD: Okay. So one of the things that we keep hearing about industrialized construction is the adoption of manufacturing techniques or approaches to construction. So what does that mean to adopt manufacturing approaches? Construction? What does that mean? And it seems like based on the organizations that we’re talking to, it has many different meanings.

[00:03:01] H.J. James Choo, PhD: Some say it’s moving work offsite into a factory. Some say it’s standardizing products to minimize design duration. Some say it’s adopting supply-led design. Some say it’s to design one and make or build many. Some are saying it’s actually change from an engineer to order, to made, to stock or made-to-order, and some are saying chunking products into modules.

[00:03:27] H.J. James Choo, PhD: I’m sure there are many other, different alternatives or perspectives that’re out there. Okay. However, one of the critical questions that people seem to be asking when they’re thinking about this is this: how do we do what and where? If we’re going to move the work from the site onsite to offsite, what does it look like?

[00:03:52] H.J. James Choo, PhD: Are we going to move the welders and the laborers actually into equipment, into the fabrication? And is it going to be craft manufacturing? Are we going to start investing in large robots to go do that? And if we’re going to start doing fabrication or assembly offsite, what are we buying versus what I’m making?

[00:04:16] H.J. James Choo, PhD: How do we make that right? Trade off decisions. Okay. So the way people seem to be going about this journey is as follows. And Todd actually again hinted on this a little bit earlier. There is a conventional design and procure and build method that everyone’s very familiar with. And in order to get to the next step, it seems like there’s a lot of emphasis on digitalization and automation of the crime processes.

[00:04:44] H.J. James Choo, PhD: So, number one, things that we’re seeing are, you know, putting in systems to automate the shop drawing processes, automate them, and facilitate the document exchange to reduce latency and document exchange and so forth. Okay. And the automation could be workflow automation through using systems or automation by buying robots, or investing in robotics.

[00:05:08] H.J. James Choo, PhD: So we’ve seen companies that invest in robots, and this  may have been talked about before, they might be producing something faster that they may not need. However, that seems to be a simple next step that a lot of people are taking. As that’s actually happening, people are starting to say, okay, we’re going to move the work off.

[00:05:27] H.J. James Choo, PhD: And that could be just a fabrication as well as that could also be an issue of assembly. One of the things that people are finding out a little bit later is, when that’s occurring, is the associated cost of additional handling, holding and cash tied up. But that seems to be something that a lot of people are going through as it moves more into manufacturing opportunity, which is fabrication and offsite assembly.

[00:05:50] H.J. James Choo, PhD: people are starting to divest into, or investing into standardization and prioritization. And the FMA that Amy also talked about, okay, once the product is being designed, and you could say standardized and prioritized, then the question comes, okay, we need to produce hundreds of these, or thousands of these.

[00:06:13] H.J. James Choo, PhD: We’re going from engineer to order to made stock, which means we need to carry an inventory of these. How do we make it? From an ordinary perspective, where is the supply chain that can make this? So what’s been happening is that the concept of “how do we go make this?” comes a little bit later than what we’re going to make.

[00:06:36] H.J. James Choo, PhD: And then once that’s addressed, then you’re getting into the more of an industrialized construction view. Now, beyond that, not sure what the road for the roadmap holds for many of the companies that are on this. But that seems to be the journey that they’re taking. And if you overlay the five levers that Todd talks about this morning, it seems to be something like this.

[00:07:00] H.J. James Choo, PhD: There’s a lot of emphasis in what we’re making, which is the product design. And then it starts to go a little bit extra into the process design element. And then once those two are pretty much well defined, then comes the latter, which is the capacity inventory and variability. Again, going back to the presentation before.

[00:07:18] H.J. James Choo, PhD: And then if there isn’t a capacity or the production system to support the product design and design, what ends up happening is going into iteration to come back to now change the product design then back and forth. Okay? So what we’re going to talk about later on in this session is maybe there is an alternative map that allows us to get, allows us to achieve the value that we’re looking for a little bit faster.

[00:07:48] H.J. James Choo, PhD: But before we get there, what we’re going to do is invite experts to talk about some of these actual elements that they’re implementing and what they’ve been experiencing in their own organizations. Okay. With that, I’m going to be handing off to Bridget Proulx to talk about her experience at Autodesk.

[00:08:10] H.J. James Choo, PhD: Bridget, can you turn on your audio?

[00:08:14] Bridget Proulx: I am on audio, but I— Oh, I can start my video now. Perfect.

[00:08:20] H.J. James Choo, PhD: We see you. All right, so I’m going to give you a brief introduction about Bridget. Bridget is a principal business consultant in industrialized construction at Autodesk. She is an architect by training with a specialty in industrialized construction, and is teaching client strategy, engagement, and implementation of industrialized methods of construction across our project lifecycle.

[00:08:45] H.J. James Choo, PhD: Bridge joins Autodesk from a general contractor, Kitchell Construction Contractors, where she spent six years working across disciplines including designer trade, partner, manufacturer, and general contractor to to leverage BIM and initialized method of construction. She received a bachelor’s in architecture and was hired by Kitchell to help define and implement

[00:09:07] H.J. James Choo, PhD: their industrialized construction initiative and challenge industry norms. Bridget, thank you, and I’m going to stop sharing so that you can share your screen.

[00:09:25] Bridget Proulx: All right. Well, thanks for having me. I am basically going to be kind of adding on to some of the topics that Amy was talking about, and specifically what I am working on alongside Amy and all of our other consultants within Autodesk in order to help our clients along this transformation framework for industrialized construction.

[00:09:50] Bridget Proulx: So with that, let’s jump into this. Hold on. Let me just make sure that I have, can you guys see my…

[00:09:59] H.J. James Choo, PhD: Yes, we can.

[00:10:04] Bridget Proulx: All right. Sorry, one second. I have to stop sharing because it is being a little buggy all of a sudden. Alright.

[00:10:16] Bridget Proulx: Oh, there we go.

[00:10:24] H.J. James Choo, PhD: Okay. We see your screen. Okay.

[00:10:28] Bridget Proulx: So one of the things that we’re doing, like you had mentioned, is we’re trying to rethink how and what the information that is flowing across the phases of the building lifecycle are, what is being influenced through manufacturing informed design. So with that, we have actually kind of refined what the roles are doing within these phases.

[00:10:51] Bridget Proulx: So as you can see here, we have operate defined products informed. Make and assemble. As operators or owners, you are articulating the end-user and operational requirements and desires in the defined products. You are the makers and you are enabling dynamic product definition for manufactured assemblies

[00:11:10] Bridget Proulx: in the built environment. As informed designers, you are setting the detailed parameters of assemblies in their aggregate, as the building requires both products and bespoke elements. As makers, you are manufacturers or builders manufacturing the parts to be assembled using advanced manufacturing processes, technology and techniques.

[00:11:31] Bridget Proulx: And as assemblers, you are system integrators. You are efficient and accelerated in integration and commissioning of manufactured assemblies within tissue construction. So the reason we kind of redefined some of these things is really just to level-set that within this convergence we are seeing.

[00:11:49] Bridget Proulx: Clients coming from the manufacturing side that want to not only make, but also assemble, we’re seeing owners who want to own that entire process from operate through to assembly, also owners who want to playbook to scale and recycle and reuse these processes and data. Manufacturers who want to manufacture and integrate.

[00:12:09] Bridget Proulx: I already mentioned that one, similarly, general contractors that are trying to become more like manufacturers and use those methodologies so that they can be more efficient in their businesses. And additionally subcontractors that are defining these products that are manufacturing and also assembling.

[00:12:27] Bridget Proulx: So these are just kind of the new roles that we are seeing that our clients are fitting within different use cases. Autodesk is hoping to help leverage those different use cases and their individual business journeys through this transformation framework. So the transformation framework is really—

[00:12:49] Bridget Proulx: we are meeting our clients where they are at within this maturity journey of industrialized construction, and we are helping them lay out that those capabilities that they need at each step to get them to that next step for that future business vision or the future state of industrialized construction, which, as you can see at the top, is this digital and physical reuse for all, for sustainability.

[00:13:12] Bridget Proulx: For, as Amy had mentioned, not re— not designing more than. So that’s where you really get your value of, you know, your business efficiencies. And so that’s what we’re working towards. So really briefly, not to go over this again, but foundationally, we are partnering with you to assess the maturity of the current IC readiness as far as data, tools, technology, culture, and really just setting that foundation so that you can actually adopt these next steps efficiently and sustainably.

[00:13:47] Bridget Proulx: Kind of beating down the wrong path. So next would be productization and ensuring that what is your product and the processes and workflows and the technology that is associated with those workflows, and we help define that and work with you to find where those gaps are and help leverage those through the digital.

[00:14:10] Bridget Proulx: So how are these workflows automated? How can we eliminate some of these processes that are redundant? How are we leveraging tools to help connect the workflows across industries and partners? Next would be the connection. How many of these workflows are being moved to the cloud to be shared with the stakeholders?

[00:14:29] Bridget Proulx: And how can we at Autodesk help connect this beyond just the stakeholders, which would then put us into optimization? Now that you are connected through the cloud, we can help leverage other opportunities with the data such as generative design, digital twins, and true industrialized construction.

[00:14:47] Bridget Proulx: Finally getting you to that point of circularity. So the things that we are focusing on are enabling that business transformation by helping our customers create that business value in applying the manufacturing techniques, technology and processes. We are connecting platform work. Implementing and capturing these workflows across the Autodesk platform and Forge services, understanding any enablers and constraints, and finally expanding the platform capabilities, partnering with customers on changing business processes to enhance capabilities using Forge ecosystem and consulting partners.

[00:15:24] Bridget Proulx: So really quickly, I just wanted to show a little clip in here of an example of essentially the consulting services that we are bringing from the manufacturing side, from this solution architecture side, from the architecture, engineering, construction, in the integrated factory modeling space, of how we are leveraging all these different processes in order to really enable that next step or that future state of manufac—

[00:16:06] H.J. James Choo, PhD: Bridget, the audio’s not coming across from the video. (Well, then you)

[00:16:10] Bridget Proulx: know what? That might have to be skipped. I apologize for that. Essentially what we are talking about is connected work. Oops, one second. Can you see my,

[00:16:26] H.J. James Choo, PhD: (see my screen?) Yes. We see your screen.

[00:16:29] Bridget Proulx: So what we’re really talking about, and in summary of what that video is supposed to be showing, is we’re talking about these connected workflows from the, from make, that is informing, helping inform design, which is helping inform the defined.

[00:16:41] Bridget Proulx: We are taking these tools, and not necessarily the specific tools under the Autodesk Cloud, but the roles of those tools and taking the data and defining those constraints, leveraging it through Forge so that you have that single data source. And that is what is really going to be connecting the engineering design and make work.

[00:17:08] Bridget Proulx: So we are partnering to help define those requirements and constraints, connect that data and provide that interoperability of those tools utilizing the platform. Another example is that we are helping leverage the appropriate levels of detail of that data to be generated and consumed between the product engineers, designers, and makers.

[00:17:33] Bridget Proulx: Oh, so once you get digitized and connected, we are leveraging our extensive manufacturing experience in our portfolio to support the advanced and emerging manufacturing processes, such as generative design, additive manufacturing, factory simulation analysis, tooling, and robotics at scale. Just as some of the examples and what some of those also look like would be

[00:18:00] Bridget Proulx: as shown, but really what the moral of our story is, And I apologize for the graphic on this because I was searching for it last minute, but I think it’s a really important slide to show where the industry convergence is happening. And the biggest problem of everyone is understanding where to really start.

[00:18:21] Bridget Proulx: And that is where, at Autodesk Consulting, that transformation framework applies to the different use cases. Of our clients and their individual journeys within that maturity matrix where we can help everyone become system integrators, understand the productization and become product makers. But at the end of the day, Autodesk is really just providing the platform for that connection of all of that content.

[00:18:47] Bridget Proulx: And with that, I will hand it over to questions, comments.

[00:18:56] H.J. James Choo, PhD: Okay. Well thank you Bridget. I think what we’re going to do is, because this is all an integrated topic that we’re going to be discussing, what we’d like to do is hand over to the next speaker to hold the questions, as we talk about time buffers at the end of the session. Then we can end the questions and everybody can chime in.

[00:19:15] H.J. James Choo, PhD: Okay. Right. So with that, what we’re going to do is now hand off to Keith Magowan from BP. Let me actually introduce him. Keith is the central subsea unit leader for BP, responsible for the delivery installation and life of field service for all of BP’s subsea production systems globally. Prior to his current role, Keith has held numerous major capital project leadership positions, predominantly in the North Sea, Egypt and Angola.

[00:19:58] H.J. James Choo, PhD: Keith is a fellow of IEE and is a graduate from the University of Edinburgh with a honors degree in electrical and mechanical engineering. Okay, Keith. Yep. Over to you.

[00:20:13] Keith Magowan: Alright, let me…

[00:20:15] H.J. James Choo, PhD: Yep, we see your screen.

[00:20:17] Keith Magowan: Very good. Thanks. Yeah, look, I appreciate the invite to join you today. It’s a topic that’s, you know, I’ve been working on for a while, but, you know, certainly this year seems to have taken a bit more of a turbo boost for me.

[00:20:39] Keith Magowan: Yeah, so starting with the energy transition, I think the first bit is the energy transition’s real. You know, there’s a lot of government and societal pressure. You know, you don’t have to go too long in the news before you’re saying something about whether it’s COP 26 for the ESG agenda or global warming.

[00:20:59] Keith Magowan: But, you know, the, I guess the companies like BP are actually responding to that. You know, Bernard came out with a new strategy when he took over as CEO in February 2020. About what BP is going to do with regards to the energy transition. So from moving from an IOC to an IEC, part of that’s going to be, you know, spending 10 times the amount on renewable energy by 2030

[00:21:31] Keith Magowan: than we did in 2019. But on top of that, we’re going to be reduced. Daily production by 40% by 2030. And that’s to try and aim towards, you know, the net zero carbon ambition by 2050 on top of, you know, the increased spend in renewables. We’ve also, you know, we’ve reduced the capital frame that we have available for us.

[00:21:57] Keith Magowan: So ultimately that means there’s less money for oil and gas projects inside of the new company. What does that mean for sub? So I’ve been looking at that with less money available for oil and gas. That means the economic hurdles to get, or, you know, subsidy projects approved, has gone up.

[00:22:17] Keith Magowan: It’s gotten harder. And ultimately that means that if we keep doing what we’re doing or what we’ve always done, it just means that subsea’s going to look very unattractive, a tight risk. You know, it takes ages to actually go from a, you know, a discovery through the end of First Oil. And when you’re competing with, you know, drilling for gas in the desert, the Permian Basin, you know, like I said, if we keep doing subsea the way we’re doing it, then, you know, it’s just not going to be an attractive proposition.

[00:22:51] Keith Magowan: So I’ve been developing the subsea vision for BP, you know, and I think three areas that we felt were really important for us was a 20% reduction in total cost of ownership. You know, and that’s trying to really start looking at, you know, higher systems and subsea systems performing across our full lifecycle rather than what’s cheapest to buy upfront, which is kind of the way we’ve always been doing it.

[00:23:17] Keith Magowan: So really looking to boost, you know, a subsea production system, a liability above 98%, so that we’re getting more out of the assets that we have available. That also then helps reduce the number of interventions we’re doing through time, which then helps the low carbon ambition that the company’s laid out.

[00:23:37] Keith Magowan: Big challenge about that, you know, and driving reliability and quality. The system is ultimately about the contracting and how we work with our suppliers. When you’re looking at return on investment, like I said, it takes ages for, obviously, projects to get designed, delivered and brought online.

[00:23:57] Keith Magowan: You know, I really want to look at, you know, taking 50% out of the defined execution of subsea projects. So, you know, we’re not investing heavy amounts of money for that long before we start seeing return on investment. And all of that’s going to be looking at productization and the delivery of speed, and then forecast and accuracy is hugely important because of the amount of money we’re spending.

[00:24:21] Keith Magowan: If we can get the forecast and accuracy correct on, you know, not only like the overall cost of what we’re going to spend between plan and actual, but the phasing that’s going through that, that means that we shouldn’t be tying up money that we don’t need to be, which then actually means that there’s more money for other projects to get invested.

[00:24:41] Keith Magowan: So what are the initiatives that we need to, you know, that’s going to help unlock the vision? Some productization, moving from engineer-to-order to configure-to-order, really getting into supplier-led design, and truly embracing supplier-led design and making sure that there’s a common understanding of what it means.

[00:25:00] Keith Magowan: And then, like I said, looking at commercial models and really trying to align the suppliers, the contractors, you know, so that their business goals align with our business goals. We look at productization, so at the moment, subsea projects are pretty much done at the top line. It’s very project specific, it’s very much operator standards that, you know, we’ve built up over time and that leads to fully customize

[00:25:31] Keith Magowan: subsea products and production systems. It costs a lot of money. It takes a lot of time and, you know, it’s a real problem for the suppliers as well, about, you know, having to continue to change tilling, change the water manufacturing, rather than actually setting their factories up for more efficient delivery of better quality products.

[00:25:53] Keith Magowan: Moving down into configuration two. So we’ve really been looking at this for probably over 10 years now, and we’ve been making steps towards it. The first step is to actually take, you know, take subsea projects and subsea equipment design away from projects. We set up a centralized group of experts to, you know, design and deliver the subsea hardware.

[00:26:17] Keith Magowan: Part of that journey was obviously, you know, trying to wrap our arms around what we were doing, but in the last couple of years, we’ve really gotten configured to order. So standard product catalogs, you know, are really kind of looking at: can we just design it once and then just use it over and over again in many different applications?

[00:26:39] Keith Magowan: We’re working heavily with, you know, the main strategic suppliers that we have. They’re all in favor of it. They’ve all developed their own kind of electronic platform where, you know, we do the pre-engineering and agreement with them upfront that then ends up in their catalog. And then when we want to start looking at new project concepts, we work with them on, you know, using the catalogs to help build. You know, it’s like Lego blocks, about what we want to, you know, what we want to use in order to actually build up new concepts.

[00:27:10] Keith Magowan: It’s been pretty successful so far. There’s a long way to go. But you know, the earlier results are certainly helping to deliver the vision that I’ve said about earlier, supplier-led solutions. I think it’s a term that’s grossly misused and misunderstood, to be honest. You know, I guess some people interpret supplier-led solutions as, you know, whatever the supplier wants to give us, it’s fine.

[00:27:39] Keith Magowan: It’s not the case, right? You know, for us, it’s really about looking at risk. Right? What’s the risk that we’re actually trying to look at? Well, we’ve accepted that, you know, the engineering standards that we’ve built up over time through experience are pretty much full plated.

[00:27:57] Keith Magowan: The suppliers on the other hand are actually looking at, you know, industry specs and what they can do with the industry specs and their own internal experience. And I think supplier-led solutions are really about that middle partner where the standard product services catalog, where we’ve actually gone and looked at and challenged our requirements versus the, you know, what the suppliers are working with.

[00:28:19] Keith Magowan: Can we accept the risk or can we not? And where we can’t, then, you know, we have that discussion, debate, with the suppliers and actually then come up with something which is mutually agreeable, both with them and us. You know, potentially it could be actually resulting in a, you know, a changed industry code.

[00:28:36] Keith Magowan: But the first step is about getting that alignment with us. So it’s not a blank, you know, we’ll take anything, but very much risk-based decisions, you know, mutually agreed between ourselves and the supply chain. Aligning incentives. So, it sounds really simple. When you say it out loud, the contractor wins and BP wins. I think it was that easy.

[00:29:00] Keith Magowan: It would’ve been done decades ago, but you know, fundamentally, the whole industry’s been built on, you know, suppliers build us equipment, we install it, it breaks, they come out and fix it, we pay them money. Right. And, you know, the cycle just kind of continues for the next 20-odd years. When you look at the contractor wins, BP wins, the full alignment of, you know, the supplier and BP’s goals,

[00:29:28] Keith Magowan: you know, that really means we need to start looking at commercial models and contracts and actually how we work together. So, you know what’s going on to get both BP and the supplier aligned. You know, I looked at this a while ago with regards to

[00:29:47] Keith Magowan: a contract service agreement. So similar to an airline engine, can I apply that into subsea? And you know, I kind of went round and round in circles with this one. And ultimately the epiphany I had was, subsea is not an airplane engine. It’s more akin to a satellite. It costs you a fortune to build it.

[00:30:03] Keith Magowan: It costs a fortune to put it where it is and it costs even more of a fortune to go and fix it if it breaks. So, you know, setting up an agreement where you’re actively trying to, you know, actively trying to maintain equipment, obviously isn’t a good idea. So, you know, it’s going to then be about pay for performance.

[00:30:20] Keith Magowan: You know, how do we make sure that what goes on the seabed actually continues to work? You know, the suppliers, a large part of their business is their aftermarket. Service is part of their business. So they’re not going to willingly give that up. So, you know, we need to look at remuneration and actually, you know, how do they get paid for every day their equipment works, right?

[00:30:43] Keith Magowan: And actually, therefore, when their equipment’s working, we’re earning money. Therefore they should be earning money when it stops working. If we stop earning money. And so today, the reaction time to fix it, you know, should improve. And actually there should be a mutual benefit about driving higher quality and more reliable equipment through this, you know, through a subsea.

[00:31:09] Keith Magowan: So when I did a presentation at the ECC forum, I worked with both SPS and Chevron. And we came up with this diagram about modern production or intelligent project production. What we figured out was, you know, the work I’ve been doing on productization is very much that left hand vertical leg, you know, about the products and the standards I’ve been working on at Chevron, and had very much been working on in the intelligent production platform without, you know, a good idea, would be to actually bring the two

[00:31:43] Keith Magowan: and, you know, actually that should be what a substitute project should look like or what a project should look like. The more I listened to that, the more I was intrigued about it, you know, following the forum. I studied with SPS about how I actually go and do this for subsea? So how do I get the process map and the plans?

[00:32:04] Keith Magowan: Because if we’re not going to be drilling in new areas, we shouldn’t be building big new projects. And it’s all going to be about fast pace, time max into, you know, big existing hubs that we have. So it should be highly repeatable, you know, and highly predictable. And I felt that was really important to, you know, go after.

[00:32:26] Keith Magowan: So for me that was, you know, looking at integrated intelligence with subsea imagined productization of intelligent products that, you know, we can build with the standard processing. The machine learning that goes on there and then actually linking that all together. And that’s going to involve, you know, a lot of work with the suppliers, because they own the designs, they own the products, but, you know, we own all the operational data.

[00:32:52] Keith Magowan: And then, you know, that should influence the design. But it’s about having that flow with that and that learning going background. So the, you know, the catalogs update as we need them, you know, as we’re starting to see things, the costs, the schedules that are being predicted, you know, upfront, you know, are believable.

[00:33:11] Keith Magowan: And to be honest, the first project that we used the productization on, I got asked numerous times about that. Right? Because they’ve never seen a project that we could actually execute at that speed, and it’s because they’ve taken all the engineering and moved it off the project.

[00:33:32] Keith Magowan: So looking at the progress we’ve had to date, I would say it certainly has been a journey for us. You know, even in the few weeks that we’ve been doing this we’ve started developing a map, we just picked a top side integrated control system because we felt there were lots of stakeholders and players in this.

[00:33:51] Keith Magowan: You know, this map probably took us about three weeks to get. In the last couple of weeks it’s expanded greatly, but it’s starting to really kind of show us that one, our processes are hugely complicated and inefficient. Two, we didn’t understand what our processes were, you know, so actually that then impacts the reliability and predictability of the projects we’re trying to do.

[00:34:17] Keith Magowan: I think, you know, if we keep going with this and we keep, you know, really kind of asking a test in every box, I think what we’ll end up with is, you know, a complicated picture that needs radical simplification. It was only the start of the week that we’ve actually fully mapped out the document control process.

[00:34:36] Keith Magowan: And you know, the conclusion from that was our engineers spend twice as much time pushing paper around through a document control process as actually understanding and managing risk in the project. Now, for me, that’s completely the wrong way around. So, you know, we’re going to be looking at what we can do to make sure the engineers are looking at the right work rather than, you know, trying to manage, I guess, an inefficient and complicated process that shouldn’t exist, I think, at that point.

[00:35:05] Keith Magowan: I’ll leave it there, James.

[00:35:09] H.J. James Choo, PhD: Thank you, Keith. It’s a very interesting journey that you’re taking and it’s a huge responsibility to take an organization as large as BP and its supply network and try to introduce a new way to do things. It’s also very interesting that you talk about supplier-led and a lot of the misunderstanding around it, and it seems like it’s very in line with what we’re going to later be talking about in supplier’s production system or capability as a starting point, because there’s no reason to set a bunch of standards that no one can go build.

[00:35:46] H.J. James Choo, PhD: Right? Yeah. And then also to actually talk about, you had three levels, which is an operator, standard contractor, standard industry. And I think this morning someone also talked about standardizing the non-customer facing part. But when you look at it from that perspective, everyone is a customer of someone else.

[00:36:09] H.J. James Choo, PhD: Yeah. Right. So, the tool coming together is pretty interesting. So we can talk a little bit more about that after other speakers talk, and then we can come together for some questions and answers. But thank you very much for your presentation. Okay. Next we are going to have Jan give us his perspective and his learnings about actually seeing a digitalized construction in practice.

[00:36:43] H.J. James Choo, PhD: So one of the things that’s going to be interesting for us to take a look at is, when we’re starting to move work offsite, it creates additional complexities on how we now start synchronizing what we’re making and delivering to what we’re installing. Again, if that’s not done correctly, there could be a huge amount of cash tied up in that queue in between, or stock in between, resulting in detrimental effects for many companies.

[00:37:13] H.J. James Choo, PhD: So, Jan, please take it away. Thank you. Yeah.

[00:37:16] Jan Koeleman: Awesome. Now I just need to make sure that you, y’all are seeing the right screen.

[00:37:24] H.J. James Choo, PhD: And by the way, I’m skipping introducing you because it was introduced earlier this morning. (Thanks,)

[00:37:30] Jan Koeleman: James. So this is going to be a bit of a different presentation than before, it’s really about the practicalities of when we choose to pursue industrialization and the realities of when that hits the site. I just wanted to share some lessons learned that we observed from a major chemical plant construction that

[00:37:59] Jan Koeleman: we are supporting. I thought that might be interesting for you to at least see. Okay, so let me take you through— I realize there’s a bit of worry, but this contains the full story that I’m going to tell. Okay. So basically we supported a large chemical plant in the ME region, which was being constructed on eMCP basis.

[00:38:21] Jan Koeleman: The civil, concrete and the structural steel were being stick built at site. But the structural steel, which is the final piece, was industrialized. And so far as that, they prepared what they called flat packs on trailers, which likely aligned with a physical space or physical 3D volume that was supposed to be put together at site.

[00:38:49] Jan Koeleman: Okay. So they had all these flat packs ready to go, and they literally had a massive laid on area full of these things. What was also going on was that the structural steel subcontractor had aggressively bid the work. Let’s say they were, they bid at X man hours per ton of steel. The productivity challenge that they were facing was that it was taking three X in terms of the amount of hours of steel installed.

[00:39:16] Jan Koeleman: So the subcontractor was facing significant productivity challenges, was facing significant losses, and was also costing significant schedule delays. Okay. But the original intention was that we have these packages, we have these flat packs. It’s going to be all smooth and dandy. Okay. When we then looked at the complications that were happening at the site, and I’ll show you in pictures in a minute, but what we started to see was that the negative variability kit site was actually fully reversing.

[00:39:49] Jan Koeleman: Despite productivity gains that they wanted to get from the flat packs, what were some of the drive factors driving that variability? So there were dimensional control issues between civil and structural steel. So they were chosen for bolted connections, and the embedded plates in the concrete were not always aligned with where the bolt holes were in the structural steel.

[00:40:09] Jan Koeleman: There were shifting execution choices. So the foreman and the people were choosing to erect differently than how the flat flags had been prepared. And there were tolerance issues, just alignment issues where the subcontractor wasn’t able to meet the specs needed now. So those were the drive to negative variability.

[00:40:29] Jan Koeleman: We then also show that the design choices, in this case, the bolted connections, really reduce the ability to deal with dementia control issues in the field. Okay. So they had to either, you know, have holes in the steel, or they had to somehow change the embedded plates. So the design choice designed for constructability, as we’ll later call it, right, had made, in this particular case, a negative variability, worse.

[00:40:59] Jan Koeleman: Okay. And then the other complication that we had was that the slow productivity was basically fully straining the relationship between the owner, the Siemens subcontractor. So the learnings that we took from that, when it comes to industrialization, right, is that if you want to capture the potential from this, and you have to deeply understand the potential sources of negative variability on site, right?

[00:41:22] Jan Koeleman: And the unintended consequences of it also, because if you understand where it can go wrong, you can understand how that will impact how many of these industrialized units you put in inventory because you’re trying to do it in such a way that you protect your work site, but at the same time, also not overproduce because that would just (a) only tie up cash, but (b) could cause significant rework on stuff that you thought you’d fabricated already.

[00:41:58] Jan Koeleman: Yeah. And several capabilities from industrialization also were clearly the value of increases significantly in an industrialized construction world, right? So obviously production planning and production scheduling, that increases predictability. It was already important in the traditional stick build world, in the industrialized construction world, you can’t do without it.

[00:42:26] Jan Koeleman: The second point that becomes extremely important in quality and station is, as I call it, it’s a term that was coined by Toyota, but it’s really ensuring that no defects are passed on, right? Because as soon as you pass on a defect, it messes up your work package that you thought was going to work and no longer does.

[00:42:47] Jan Koeleman: And if you then have partial erection of a work pack, you can imagine what that does. And again, I’ll illustrate in a minute. And the third thing is that design for construction. So really knowing how your product design impacts your process and your execution, right? That also becomes a capability that’s extremely valuable.

[00:43:10] Jan Koeleman: And then lastly, in aerospace, that’s called production engineering. I think in this case uh, we’re talking about the process design as PPI represents it. And then last, you have the load balancing between fabrication and insulation is critical. Again, it’s the optimum between how much you have at site, eh, to be able to have an alternative workfront that the primary doesn’t work, but at the same time, not too much that you risk the rework that I mentioned before.

[00:43:43] Jan Koeleman: Anyway, that’s the story, in short, I was just going to illustrate. So firstly, the project that we were looking at, right? Gives you a bit of a sense of what it looks like, the stage it was in, right? So Civil Works had already been done and they were starting to do the structural steel around it, which you can see here.

[00:44:04] Jan Koeleman: Okay? Now,

[00:44:08] Jan Koeleman: and this project was, in order of magnitude, somewhere between one and two U.S. dollars.

[00:44:18] Jan Koeleman: The issues we observed, which is how we got to the learnings, was when, so as we said, we were brought in because there were productivity challenges being experienced by the subcontractor impacting the schedule. So we went in and we did some field based observations. What did we see? We saw, for example, significant double and triple handling of material.

[00:44:38] Jan Koeleman: So the cranes, which were at the location of work that were intended to be used to erect steel and lift steel into place, were being used to conduct multiple picks to offload and reload pieces coming from the flat pack. And I have an image in the next page that shows you what we mean with the flat.

[00:44:59] Jan Koeleman: What we’re also seeing was that there was our sequence work, right? So elements were unable to be finished because of the quality of work. So we had pieces of structural steel that were left hanging as a result of the embedded plate being in the wrong location. We also observed, for example, Rens time losses.

[00:45:15] Jan Koeleman: So there was a situation where the wrong piece of steel was picked, and it was probably because the inventory at site was too much. And it took them 45 minutes to figure out that it was the wrong piece and they had to replace and get the right one. So issue resolution was not so relevant for this discussion.

[00:45:34] Jan Koeleman: But anyway, it’s one of these examples where the wheelchair wiper of a crane was being used or it hadn’t been repaired in weeks, and the subcontractors who were needed to manage this together hadn’t aligned it. But that’s a separate thing. But the other thing that was interesting was the lean mindset.

[00:45:52] Jan Koeleman: So on this piece of steel, that was the wrong one. After 45 minutes they figured it out and resolved it, and the crane lifted the right piece into place. We engaged the foreman to ask him what needed to be improved in this particular situation. And he said, we need faster cranes. Now, as you might imagine, that’s not exactly what we would argue is the right solution, but it very much showed the traditional mindset.

[00:46:20] Jan Koeleman: You know, predominantly wanting to focus on additional capacity, so more crane capacity, faster crane capacity, but also the inventory increasing the LA down area. Okay. So when we look at the root causes of some of these, and let me highlight the one on the next page. So the idea of industrialization was noble, right?

[00:46:39] Jan Koeleman: We said, if we prepare these flat packs and the flat packs arrive, right? You’re managing, basically, or ensuring, all the right pieces are there to construct the work is required. Yeah. But when we looked at these, at the composition, either the flat packs were optimized for the fabrication or for the fabrication workshops, right.

[00:47:01] Jan Koeleman: Or for volumetric space. We couldn’t completely discern which one it was. Either way, they ended up not being the required construction sequence. And then what does that practically end up happening? So on the left hand side, you can see the flat pack. So it’s a trailer that a truck can simply lift.

[00:47:20] Jan Koeleman: All right, and then on the right hand side, you can see it being offloaded. Now what you’ll see, I don’t know whether it’s super clear, is that for a flat pack, you have the heavy parts on the bottom and the lighter parts on top. So if you’re out of sequence and you need one of the heavy beans at the bottom, you need to unpack everything.

[00:47:44] Jan Koeleman: Get the heavy seal out of it, put the rest back, and then take it away again. Okay? So that’s where the productivity losses really accrue because this is the location of work. This is where your critical production is happening and where you want to have the highest productivity, the quality issues. I don’t know whether this is completely clear what you can see here, but on the right hand side of the picture, right, you can see that the embedded plate in the back was missing bolts.

[00:48:15] Jan Koeleman: So these structural steel members were left hanging. So again, from the principle of industrialization, if you have this kind of quality issue feeding through it completely reverse. The objective of what you’re trying to achieve with industrialization, which is a very smooth, very, very consistent flow of work.

[00:48:34] Jan Koeleman: Yeah. And on the left hand side also, you can see that these beams that are on the left hand side, they’re actually crooked. So it means that the two beams, this one and that one, are slightly out. Also tolerance issues, and you can see that it couldn’t finish the rest of it. Okay. So the equality issues are absolutely killing from when you’re trying to do this industrialization.

[00:48:56] Jan Koeleman: This is what the total inventory site looked like. Here you can see again the flat pack and all the parts that they kept in place, right? Thinking that inventory would protect productivity, but which obviously enabled them to freely choose and shift execution choices, meaning that the next flat pack was even worse in terms of having the right pieces than the one.

[00:49:24] Jan Koeleman: Okay. And then lastly, as I said before, right, these bolted connections, right? Really made the negative variability, as in if there were dimensional control issues, even worse, because of all the reports of rework that needed to happen, and obviously it couldn’t be done at this location of work; it had to be moved back and transported again.

[00:49:48] Jan Koeleman: What we just wanted to share in this brief story is that, look, there’s many benefits. Industrialization has great potential, right? So, it enables product standardization as we just discussed. It enables production standardization, as we just discussed, and as I just discussed, I mean, in the previous presentations, right?

[00:50:05] Jan Koeleman: It allows you to capture the learning curve. We should do this in a more standardized way. It allows inventory optimization products to begin logistics optimization. There’s a whole bunch of things that come with these benefits, but there’s also a real need for certain capabilities, right, which include production control, including design for construction or production engineering, right?

[00:50:28] Jan Koeleman: The quality and station, making sure that no defects should pass on it. It goes into production system optimization, understanding what the optimal level of inventory is to connect fabrication with the site, acknowledging that there will always be a bit of variability, right? It requires data and analytics and inventory tracking control.

[00:50:49] Jan Koeleman: Those are the types of capabilities that, in the industrialized world, as others I’ve already talked to, will become materially more important. That’s what I wanted to share. Just lessons learned. So

[00:51:00] Jan Koeleman: I hope that was interesting from the discussion.

[00:51:05] H.J. James Choo, PhD: Thank you for sharing your experience and that story.

[00:51:09] H.J. James Choo, PhD: It’s an incredible story that, again, I think repeats the learning that even if you actually have or are trying to improve, making certain decisions without understanding implications on the overall design could have a detrimental effect, which wipes out the original intent that you’re trying to achieve, so I really appreciate that experience.

[00:51:33] H.J. James Choo, PhD: Okay. All right. So one of the things that we’re going to do, that we’ll question, was about tolerances. Now, the tolerances come from the product tolerances, right? However, there’s other tolerances that we can actually talk about, which are temporal tolerances. When do certain things need to get to,

[00:51:57] H.J. James Choo, PhD: and when do certain things actually need to get done, and that’s not a typical construction management science, but as we talked about before, it is about the science of: how do I make things? So, and that actually is the fundamental element. Let me actually see if I can go back and share my screen here.

[00:52:18] H.J. James Choo, PhD: The fundamental element of industrialized construction. So again, taking manufacturing approaches and adopting it to construction. So what we’re going to do is talk a little bit about moving away from the science of a thing. How do we design things and actually move into the science of making, okay. To do that, what we’re going to do is invite Mark Spearman to give us a little bit of an overview on what operations science is, and let me introduce him.

[00:52:49] H.J. James Choo, PhD: Mark joined Strategic Project Solutions in 2018 as a director of technical solutions, as well as technical director of the Project Production Institute. He is the founder, president, and chief executive officer of Factory Physics Inc., a firm that is now part of SPS and provides management, consulting, training and software to improve manufacturing and supply chains.

[00:53:13] H.J. James Choo, PhD: Spearman co-authored the award-winning book, Factory Physics: Foundations of Manufacturing Management, which has led a paradigm shift in the approach companies take in managing their operations. So with that, I’m going to actually hand it over to Mark. Mark, over to you.

[00:53:33] Mark L. Spearman, PhD: Okay. So I’ve been listening to these talks. I’m very happy to hear about applications of operations science and how it’s being applied. We’ve always thought it was a very general application, and it turns out that it is. So that’s very gratifying.

[00:53:57] Mark L. Spearman, PhD: Operation science takes a different approach. We use a different TE terminology. We have different measurements. Instead of talking about Gantt charts and last responsible moment, integrated work packages and so on, we’re going to be talking about work in process, throughput, cycle time utilization, fill rates, inventory investment, and all these kinds of things.

[00:54:20] Mark L. Spearman, PhD: And it’s not that project measures aren’t important. It’s just this: there’s other measures that are also important, so we have to make sure we do all of those, both of those. So I’ll go back to Todd’s slide. Traditional project management – you define the scope and quality. Talk about the schedule.

[00:54:44] Mark L. Spearman, PhD: When does it need to be finished? And then you look at the resources that you need to do it. EPM. Look at the product designed, you know, it can even be what the specs are of, the deliverable, in order to make it more useful for the ultimate customer and also make it easier to install, to build. And then we really focus on the process, the system that’s going to deliver the project production.

[00:55:16] Mark L. Spearman, PhD: As we’ve been talking about all day, production systems involve capacity inventory and are affected by variability. And so that’s kind of a big difference between the two approaches is that we look at the production system to see whether or not it does have the capability to deliver this project at the time that it needs to be delivered.

[00:55:40] Mark L. Spearman, PhD: And the way we do that is we first build a simple map of the process, looking at all the steps that need to be done, and we convert that map into a model. The model could be an analytic model. Usually we start with an analytical model, and then we also can do a Monte Carlo discrete event simulation model.

[00:56:05] Mark L. Spearman, PhD: I’ll talk a little bit more about both of those. Once we have a model, we can simulate the production system that will deliver the project. We can analyze it and make sure there’s enough capacity, there’s going to be enough work in process, not too much work in process, enough inventory, the lead times are going to be appropriate and so on.

[00:56:23] Mark L. Spearman, PhD: And some things we can actually optimize. Two things that are really important to optimize are the inventory policies. When do we need things to be on site so that we will not delay the project, but not so early that we clutter the site. I mean, we’ve had things like a distillation column delivered two years before it was needed.

[00:56:46] Mark L. Spearman, PhD: And you know, this is not good because probably the requirements for that column will change in those two years. Not to mention the environment it was stored in isn’t good for it. So we want things, we don’t want things delivered just in time necessarily. We want ’em so that there’s a low probability of not having it.

[00:57:07] Mark L. Spearman, PhD: Very low probability, and we’ll show that where a lot of companies have no probability of being late, but they have way too much inventory. Also, once we optimize those parameters, we use those parameters to control the project so it feeds into the control systems, tells us when to bring on more.

[00:57:33] Mark L. Spearman, PhD: When to release more jobs, when to order all these things. There’s thousands of different parameters that have to be optimized. So you need to have a system that can do a lot at once. And while we’re executing the project, we’re looking for ways to improve, and when we’re doing that, we’re collecting more data.

[00:57:55] Mark L. Spearman, PhD: So those improvement ideas and the data go back into the map of the model and improve it. Make a better map, a better model, continue to analyze and optimize and continue to control and improve. So it’s an iterative cycle. So the focus in traditional project management is milestones in time given by a Gantt chart, and

[00:58:20] Mark L. Spearman, PhD: the idea is that if you look like you’re going to miss one of these milestones, you go and make some adjustments. You add capacity or reprioritize or do something. And then on the right hand side here, we see things like, you know, how much work is done, how much work’s been planned, how much have we gotten paid?

[00:58:43] Mark L. Spearman, PhD: These kinds of things people worry about, but these measures are important, but they’re not looking at the production system. So the difference between the digital twin that runs the old way of doing this to the new way of doing this is the new way. The focus isn’t on milestones in time, but target rates, target production rates, and rates are directly related

[00:59:11] Mark L. Spearman, PhD: because the rate at which tasks are completed can never exceed the capacity of the project. The capacity of the project is the bottleneck of the project. But if you maintain the rates, you will hit the dates. So that’s a very key thing that we focus on. And rates are something that we look at every day.

[00:59:31] Mark L. Spearman, PhD: If we don’t look at it when we start, you know, our rate falls. We look at it every day and make sure that we’re maintaining the rate that we need to maintain. At the same time, we can do some very quick analysis. We call this absolute benchmarking, and basically it takes four pieces of data. My current WIP, my current throughput, throughput is basically how many tasks I need to do and how much time divided by how much time my current

[01:00:03] Mark L. Spearman, PhD: bottleneck rate, and the raw process time, which is the time to go through the complete task if there were no queuing or interruptions like that. And we can plot this on this graph, and what we see at the very top, the current state, the green line is the throughput. Blue line is the cycle time.

[01:00:28] Mark L. Spearman, PhD: The top green line is the ideal. The dashed green line is what we call the practical worst case. Likewise, with the solid blue line is the ideal and the dash line is the practical worst case. In this case, you can see that the throughput is actually below the practical worst case, and the cycle time is above the practical worst case.

[01:00:47] Mark L. Spearman, PhD: After performing the optimization, we move the throughput into what we call the lean zone, the good. It’s above the practical worst case. It’s never going to be at the ideal. But now what we’ve done is we’ve increased throughput from around three tasks per day to about four and a half tasks per day. And the cycle time has gone from about 25 days to about 15 days.

[01:01:11] Mark L. Spearman, PhD: So it’s a huge improvement. This is a small part of a project, but this flow was improved a great deal and you can see it very easily on, very easy to do. That’s a kind of a first cut, but after doing something like that, we’ll want to have a more detailed project. And so this is a small piece of a larger project.

[01:01:40] Mark L. Spearman, PhD: You can see down on the lower right here, the rest of the project. And we also, by looking at this, we can see that there is a potential problem. All of these, these parts coming from different, some of these are different suppliers. Some of these are some of our own fabrications. They all have to come together here before we can do this operation here.

[01:02:04] Mark L. Spearman, PhD: And we call that wait to match time. And the way to, the only way to handle that is to make sure we have an inventory buffer here or, essentially, a lead time. Lead time buffer is going to end up being an inventory buffer as well. So we want to max— we want to make sure that that inventory buffer we have is big enough to accommodate all of these things coming together, but not so big that it exceeds the inventory requirements that we should require.

[01:02:37] Mark L. Spearman, PhD: So the next slide shows some of the basic equations that we use, and if you’ve attended these before, you’ve seen a lot of these. The first one is Little’s Law. It says WIP is basically proportional to throughput. ICYC time is basically proportional to WIP because the throughput is going to be the number of tasks divided by the duration that you’re allowing, so that’s fixed.

[01:03:07] Mark L. Spearman, PhD: So basically WIP and cycle time. Two measures of the same thing. The only difference is you can see WIP, you can’t see cycle time. And also WIP is a leading indicator of cycle time. If WIP goes up today, cycle time will go up next week or next month. And so controlling WIP is very important. Trying to control cycle time is kind of futile.

[01:03:30] Mark L. Spearman, PhD: You end up with an uncontrollable system that oscillates terribly on the other. Important formula is the cycle time formula which says cycle time is raw process time, plus batch time, plus move time, plus acute time, plus shift differential time, plus the wait to match time I mentioned, plus the planned time buffer.

[01:03:50] Mark L. Spearman, PhD: The raw process time is the actual time you’re doing something. The batch time includes wait to batch and waiting in a batch, so you may be accumulating a batch before you do the process, and after that process, everything’s moved to the next workstation where it can’t all be worked on at the same time.

[01:04:12] Mark L. Spearman, PhD: Move time can be long, but typically on a project it may not be that long. The longest times are queue time and batch time, and typically those two added together, if you take out the planned time buffer, those two added together represent probably 90% of the actual cycle. Not including the wait to match time or the planned time buffer.

[01:04:38] Mark L. Spearman, PhD: The shift differential time comes about when you have different crafts working on different shifts, and so things are stationary. The wait to move time we talked about on the previous slide. The planned time buffer is usually a CYA that’s put in there to make sure that we get things finished on time.

[01:04:56] Mark L. Spearman, PhD: And so we really don’t want the planned time buffer in there if we don’t need it. And so,

[01:05:02] H.J. James Choo, PhD: just actually at a point, I think is, this is a pretty, actually a deep conversation that a lot of people could benefit from, but not, not because of constraint time. Again, I think it’s a topic that we’d be actually happy to discuss with you in more detail if you’re actually interested, but one of the things you might have already been recognizing is we are speaking a different language, and that is the language of manufacturing that Mark is introducing.

[01:05:32] Mark L. Spearman, PhD: Right. Thanks, James. So, yeah. Anyway, the foot equation is also cycle time. So once we build the model, we can use analytic models or we can use discrete event simulation, and the analytic models we talked about before, and the discrete event simulation shows like an actual picture of the WIP as it goes through.

[01:06:00] Mark L. Spearman, PhD: And you can run this for different random variables and see how much it changes each time. Notice that the utilization, this, these are the utilizations of the resources. They’re basically the same. This graph is a little easier to digest. And looking at a noisy picture of the WIP versus time, the important thing is, wait…

[01:06:23] Mark L. Spearman, PhD: Okay. And then here are some of the basic graphs. Here’s what I was talking about with the inventory. So, the X axis is the probability you have a given part in inventory, and then the total investment is on the inventory. Investment dollars is on the Y axis, and I believe these are in thousands.

[01:06:43] Mark L. Spearman, PhD: So this is probably, it’s not just one part, maybe a thousand parts. Currently they’re operating up here, so about $80 million in inventory, and you can see that you could be down here with more frequent orders and a lower target and still be at a hundred-percent fill rate. So that’s a huge savings up here is discounted cash flow and shows that the benefit of

[01:07:12] Mark L. Spearman, PhD: reducing cycle time to bring in the first. This is from a well field development. So finally, there’s the make versus buy decision that needs to be made. And we can analyze this as well. Some of the KPIs for making. You have a service level, which is, what’s the probability?

[01:07:39] Mark L. Spearman, PhD: Your lead time is long enough to get things done. If you’re going to make it, you’re going to have to invest some money in capital equipment. And then there’s going to be a cost, labor and whatnot. And then you’ll have cash tied up in  WIP. If you buy, then you’d be looking at a fill rate for the inventory.

[01:08:01] Mark L. Spearman, PhD: You don’t have so much investment, but you still will have a cost. You’ll have a cost per part as you buy it. And then it’s also cash tied up in this inventory. So those are kind of the different things you have to consider. And finally, project management looks at labor productivity, float cost variance, schedule variance cost, and schedule performance indices.

[01:08:29] Mark L. Spearman, PhD: And these are still important. Maybe some of the indices aren’t so important, but we’ll still be looking at all these basic measures. But we’ll add to those service level and fill rate, WIP, what we call critical WIP versus how much WIP position you have. The capacity utilization, the back size turns out to be a huge thing, can really change the time it takes to finish the project.

[01:08:54] Mark L. Spearman, PhD: But optimizing that inventory position includes not only what’s on hand but also what’s on order. And those both need to be monitored and controlled with reorder points and reorder quantities. And these are some of the reorder points and reorder quantities and back size. Those are some of the things that we can optimize, and after we optimize, it goes into the system and then the system becomes more effective.

[01:09:20] Mark L. Spearman, PhD: So I’ll stop there and we can take questions at the end.

[01:09:25] H.J. James Choo, PhD: Okay. Thank you, Mark. So, what I think the point that we’re trying to convey is, when we’re now starting to adopt manufacturing technology techniques and approaches to construction, that it requires us to rethink some of the methodologies and approaches, but also the metrics and

[01:09:49] H.J. James Choo, PhD: the questions and answers that we might be asking ourselves. Right? So with that, what we’re going to do is now quickly hand over to Roberto, who I don’t think actually needs introduction as he’s been leading the session all day. But he, as a senior vice president of SPS technical services, has had the opportunity to work with many organizations that are on this.

[01:10:18] H.J. James Choo, PhD: And what he’s going to do is provide us some insights he’s gathered from supporting those organizations. Roberto, over to you.

[01:10:25] Roberto J. Arbulu: Thank you, James. So we are going to bring to you some ideas. What we are seeing, you know, we call it here, some field observations on organizations that are pursuing industrialized construction.

[01:10:38] Roberto J. Arbulu: And one of the questions that we should ask ourselves is, you know, industrialized construction is really pushing the project production systems and changing how they are configured, right? So the question is, are we making the right decisions about those configurations, right? And so let’s think about that question as we go through, and I think you should think about that as well.

[01:11:06] Roberto J. Arbulu: Okay, so what do we see, as James mentioned earlier? And so I’m just bringing that same diagram to keep it simple. That obviously industrialized becomes that intersection, that convergence of construction and manufacturing practices. However, what we see, if we start with one on the right, is that there are a lot of construction companies that are moving to the left, meaning that they’re actually trying to provide services for their customers and

[01:11:38] Roberto J. Arbulu: jump on this movement of industrialized construction. Right. We also see manufacturing companies that are not construction companies, just to be clear, are companies that are embedded in the making of parts, components that are moving to the right. Right. And they’re actually offering services to customers, owners, right.

[01:12:01] Roberto J. Arbulu: Different types of customers about how they’re going to start assembling. Fabricating, assembling and delivering, not just apart, but a much more complex assembly or group of assemblies, or you could even call them modules, right? And so if we also look on the right, what the industry has been experiencing for, I would say, many, many years, right?

[01:12:29] Roberto J. Arbulu: There’s nothing new about that for a given project, right? A construction company might make decisions about fabricating and assembling things off site. Right. Or even fabricating and assembling things on site. Right. In industrial construction, we see a lot of that with fiber spools, for instance, right?

[01:12:52] Roberto J. Arbulu: People are fabricating, fiber spools are cutting, bending pipe, welding pipe. And creating an assembly, right? And so this actual change in the industry is reconfiguring supply chains. And this is happening in front of us, in front of our eyes, right? And I think we need to be conscious that this is not only happening, but we need to be ready to determine once again, what are the implications for us?

[01:13:28] Roberto J. Arbulu: When I say us, I mean all the different parties in the engineering and construction industry. How do we engineer, how do we fabricate a symbol, deliver things, and how do we handle installation? Right? And so I want to bring a couple of examples to you in a minute, but what is important to emphasize is for those who are in construction, right?

[01:13:52] Roberto J. Arbulu: We talked earlier, Todd mentioned this from a heavy construction management perspective. A lot of companies we’re seeing, they’re bringing this conventional, let’s use that word construction management perspective, and trying to apply that in an offsite environment, an assembly facility that is not going to work, right.

[01:14:17] Roberto J. Arbulu: A lot of them maintain a position to focus quite strongly on an administration approach versus a project production approach. Mark just took us through operation science and, although that science has been applied and is being applied in manufacturing, on the left side of that Venn diagram, it applies to any sort of production system, right, including temporary production systems like projects as the industry changes, and moving this into industrialized construction.

[01:14:53] Roberto J. Arbulu: And so let’s take a look at a couple of examples. I’m conscious of time. I want to move extremely quickly on this. So what you see here are two representations of production systems, right, happening in the same physical facility, in the same assembly. The blue line around each of these represent the physical boundaries of this assembly facility.

[01:15:25] Roberto J. Arbulu: What you see outside the blue dotted line are the different, let’s call them supply flows of different components, different materials, different sub-assemblies that are needed for this entire production system to generate a throughput. A throughput might be a module going out of the production system to meet certain demand, right?

[01:15:48] Roberto J. Arbulu: And so what you see in an actual real life example, is a production, two production systems physically in the same facility, right? The complexity of what is outside the facility, just visually, right? I’m not running any analysis just visually. It’s much larger than what happens inside the facility itself.

[01:16:14] Roberto J. Arbulu: And so what we see, once again, we’re talking about field observation, is companies not really understanding the end-to-end production system that they are supposed to be managing and concentrated only on what they are directly responsible for. We believe that is a big mistake and the industry should be conscious about that because the complexity is quite, quite higher as you can see here in other places.

[01:16:41] Roberto J. Arbulu: The question is how much inventory are we going to allocate? What are the lead times associated with all these different components of the production system outside the blue circle, the blue area. Okay, so the first step is: do we understand the production systems that we are managing or are supposed to be managing?

[01:17:05] Roberto J. Arbulu: Number one, are we making the right decisions about how we handle the entire production? Or even a component of it, like these are here that I’m circling with my mouse, that represents, in this particular case, still a structure that goes with the module that the entire production system generates. So one thing that Mark actually took us through, and I’m not going to repeat it because he explained a little bit of this, but what you see here is actual true data from the previous

[01:17:43] Roberto J. Arbulu: production system that you saw where the current state of how the production system behaves with capacity utilization of 59. We estimated this, we calculated this using operation science that will give us a throughput that is about 3.2 and also we’ll have a cycle time that is about

[01:18:05] Roberto J. Arbulu: 20, 23 days or so to deliver a

[01:18:10] Roberto J. Arbulu: portion of that. Right. And what is important to highlight here is that capacity and resources are not the same. Using the same amount of resources, we can increase the capacity utilization of the production system to a much higher value. 85%. We are not saying a hundred percent because, as Mark and even Todd earlier introduced, the closer we get to a hundred percent, the longer the cycle.

[01:18:39] Roberto J. Arbulu: That’s not good. So we want to keep, we want to maintain that capacity utilization of the entire production system to a value around the one you see here, 85%. By doing that without increasing resources, this is very important, we can significantly increase throughput and we can reduce cycle time, which is also important in order to be meeting service levels and address the demand.

[01:19:06] Roberto J. Arbulu: If we take a much higher level view of the difference between lead time and cycle time. In this example, the one that was shown earlier, we have a lead time for a given module of about 40 or so weeks, right? 40 weeks. You are talking about 10— let’s do the math, right?

[01:19:30] Roberto J. Arbulu: And that includes being, they need to assemble a frame for or still frame and then integrate and then deliver, right? But, and all this is the high level different flows that need to be, they need to be managed by this organization. One of them is electrical gear that is being supplied by their customer, but the other ones are elements of their production system that they need to

[01:19:57] Roberto J. Arbulu: managed directly. This is a subset of just the steel production system, right, where you have, you know, going all the way back to the meal components, ship into the fabricator, different organizations involved, and this is a true story. Look at how many different parties are involved and the lead time that it takes in order for the company that is fabricating and assembling those modules to get the steel frames to be painted, assembled, blasted, and then shipped to the assembly facility.

[01:20:31] Roberto J. Arbulu: Right? We are, as an industry, misunderstanding or not fully understanding the complexity and what decision we’re making. When we say we’re going to go industrialized, how do we handle this? Right? We cannot control the execution of work with placing POS, right? This is not a commercial problem, right? How can we decouple the commercial from the operations, right?

[01:20:58] Roberto J. Arbulu: This is another example that I’ll finish with. Otherwise, James is going to get upset. This is of a manufacturing company that was moving on that band diagram from the left to the right or to the middle. Right. And this company was tasked by a customer to fabricate a specific type of modules. Right. But with an increase in demand from four modules to eight modules.

[01:21:25] Roberto J. Arbulu: And these modules are not exactly identically, physically the same. Right? The question is: is the production system able to meet the demand, right? This is something that we were able to understand: what is that production system? How it’s configured. And this company made a decision to decouple the modules, physically decouple the modules into sub-components, and give some of those components to a third-party manufacturing company.

[01:22:01] Roberto J. Arbulu: What the owner might have not realized, in this particular case, is that someone on site is going to end up with the job of assembling the module with different components. And because these different components are not identical, they need to arrive in a specific sequence. So although we’re making a decision to fabricate offsite and modularize, and even when we can actually talk about productization,

[01:22:30] Roberto J. Arbulu: not having an entire perspective of the production system, may move the industry and the players in the wrong direction because, for the guys doing the installation on site, we can see here, this is going to be a nightmare, and all the benefits of far keening of site might be lost by just dealing with this integration, physical integration in the field.

[01:22:54] Roberto J. Arbulu: Okay, so I’ll leave you with the same question. Are we making the right decisions about how we configure product production systems? Right? There’s a lot of things to learn about the confluence, the congruence of convergence of production systems, and the idea of industrialized construction. Over to you, James.

[01:23:13] Roberto J. Arbulu: Thank you, Roberto.

[01:23:15] H.J. James Choo, PhD: So, Roberto’s given a very interesting example of what he’s seeing out there of actually starting to look at the boundaries of what they created as the factory. The question is, everything else that they’re now looking at is what they buy. So one of the opportunities to figure out is: what does that boundary look like for your factory if you’re going into industrialized construction?

[01:23:39] H.J. James Choo, PhD: Now, at the start of this session, we actually talked about the typical approach that people are taking to get to industrialized construction. And the question that we posed was: is there a faster roadmap? Okay. Is there a faster roadmap? And I think what Autodesk has talked about, what VP Keith has talked about, it gives you a little bit of hint in maybe how to accelerate this process.

[01:24:06] H.J. James Choo, PhD: But we propose that rather than taking a sequential approach to focus on the product, then process, and then working on the production system, it proposes that there is a concurrent design opportunity here. So, as Todd talked about this morning, there’s no free moves. So if you’re going to change something, that has implications for something else, so why, rather than doing it sequentially, how do we do this?

[01:24:29] H.J. James Choo, PhD: Why don’t we actually do this concurrently? And when we’re doing it concurrently, the question becomes: so how, what does the actual roadmap look like in this new model? And this is what we’re proposing. Okay. We did not as a humanity, as a society, we did not wait until we had mapped new business processes.

[01:24:51] H.J. James Choo, PhD: We educated everyone, and they have changed all the business processes before we started adopting mobile phones into our daily lives and business lives. We leveraged technology to shape how the new future might look like. And I think this is exactly the opportunity that we have for construction.

[01:25:11] H.J. James Choo, PhD: So as we’re moving into industrial life construction, as BP talked about, as Amy talked about as one of the foundational elements, I think we’ll call it process, is to understand the capability of your supply network through the production system. Design, optimization, and control. Make that light up.

[01:25:32] H.J. James Choo, PhD: Talk using the IoT because there is already movement going on in your supply network. Through the industrial 4.0 and supply chain 4.0. Some of this concept I think Todd and his group will talk about in the next session. And then based on the capacity and the capability of your supply network, start actually thinking about should we actually standardize and prioritize in order to leverage or increase the capacity of your production system if it needs to be.

[01:26:05] H.J. James Choo, PhD: At which point you’re starting to realize industrialized construction, and then by adding all the AIML as well as the realtime feedback and feed forward, you’re now starting to create intelligent production. So rather than relying on the conventional tools and methodology to transform the industry, what we’re saying is leverage technology that’s already out there and approaches that the other, the industry already have benefited from, and leverage that to speed our journey through the change.

[01:26:39] H.J. James Choo, PhD: Okay. Again, that’s to get you started, the first step There are few things that we’re going to offer. One is the free tool that’s actually provided by PPI. Again, as I said, free tool. Just go to the resources and tools and process map on the PPI website to start gaining understanding as to what your process looks like.

[01:27:01] H.J. James Choo, PhD: Because as Mark proposed, mapping is your first step, and then if you want to learn a little bit more about “how do I do this right?”, there is a certificate course that’s being offered  in conjunction with Cal Poly on these three tracks. Again, you can find out more details at the PPI website. And with that I’m going to hand it over back to Roberto.

[01:27:24] H.J. James Choo, PhD: Sorry we don’t have additional time for questions, but we do have some questions and I will share it with our speakers, and we’ll get back to the people that have asked the questions. Thank you.