UK Government backs a Platform approach to industrialized construction

JAIMIE JOHNSTON: Hi, there. My name is Jaimie Johnston. And I'm going to be talking about the way in which the UK government has adopted a particular approach to industrialized construction.

My job title is Head of Global Systems at Bryden Wood. We're an integrated designer. I'm also the Platform Design Lead for the Construction Innovation Hub. And you're going to hear a bit more about at the Hub as I do my talk.

So to put some context around this, UK government-- certainly in the last six to eight months-- have been making some fairly interesting announcements about its particular approach to industrialized construction, and how it plans to transform construction from a very labor intensive, costly, carbon intensive sector into something much more like manufacturing, with all the benefits manufacturing seen in terms of time, cost, quality, carbon, productivity, et cetera.

But the announcements that I'm going to talk about and the kind of recent approach has actually got quite deep roots. So it's a journey that government has been on for a number of years, at least four years now. And I'll talk you through some of the key documents, and the way in which government's thinking has crystallized and refined over time, from the kind of early announcements around a move towards modern methods of construction to the current thinking around some very, very specific mandates.

So the obvious question is, what are construction platforms? To bring it to life, I think everyone is already familiar with the platform concept. So as you're aware, I'm sure, when you go to IKEA, it doesn't matter whether you're building a bookcase or a bed or a chest of drawers or a wardrobe. You can build a whole range of different types of furniture.

But you'll have noticed that a lot of the fixings, the knock down fittings, are identical, regardless of what you're making. The book is always in the same format. It's always very visual. It's always easy to follow. The tools that you use, whether it's an Allen key or a hex key, are always identical and very, very limited in what you need.

So IKEA has managed to take an enormous range of furniture, boil it down into a relatively small number of components. And they make those components at enormous scale. And that's how they drive down the kind of cost benefits.

They've managed to make it very easy to assemble, which is why anyone can make IKEA furniture. And it's transformed furniture industry from being a sort of like construction-- fairly trade based, relatively expensive thing-- into something that anyone can do. And I'd argue that probably, the quality of design in people's homes has generally increased as a result.

So construction's looking to do the same thing. So one of the conversations that government started four years ago were saying, well, what is the IKEA kit for construction? How would you develop one? And what would it look like?

And one of the key things that we started talking to government about was this idea of rather than focusing on the difference between sectors-- so rather than working out why schools were special or why health care was special-- if you started to say, what's common and what's similar across all these sectors, you could start to identify the sort of key features which are repeatable. That would allow you to develop the kit of parts.

And then, if you were doing those across sectors-- so rather than buy components for each individual project or program, you were buying them across the government estate-- that's how you get into the numbers that manufacturing works with. And that's how you make a transformational shift away from sort of project thinking to something much, much broader.

So this is a video that Infrastructure Projects Authority, part of the UK government, released. So I'm going to leave that video to run. And it's a very good articulation of the benefits of what we mean by platforms.

[VIDEO PLAYBACK]

ALEISTER HELLIER (ON VIDEO): Hi, I'm Aleister Hellier, Head of Benchmarking for the Infrastructure and Projects Authority. I'm going to explain in this animation how we need to change the nature of our interventions in the build environment, to address the need for social infrastructure using platform approaches to construction.

We normally do the public buildings as standalone projects. And we could deliver more efficiently if we took a different approach. If we could get a shared view of what projects are in the pipeline across the public sector, I think we could see which elements they have in common. We could use that knowledge to invest in construction platform solutions that everyone can benefit from.

Having a predictable pipeline and clear rules for platforms would give industry the confidence to invest in product ranges, products that could be configured digitally bespoke to the project context, but consistent with the rule set. It would also mean that the associated investment in manufacturing assembly, created inclusive employment, could be spread across the country with the assurance that all components will work together, as expected, when they are assembled on site.

Implementation data could then be used to feed back into the system so that learning leads to improvement, not just in a single building, but to the design and delivery of all future projects created using the same construction platform or products. This lesson will help us understand whether our investments are actually achieving the priority outcomes [INAUDIBLE] anticipated, and lead to a virtuous cycle in how we use the built environment to meet the needs of each citizen.

JAIMIE JOHNSTON: So that's, I think, a fantastic articulation of the process and the benefits. So I was going to then talk through how did UK government arrive at this point, and what were the kind of steps it took, and how did the thinking evolve over time. So go back to the start, the timeline here.

The first document that started to describe this was something that Bryden Wood wrote while working with the Ministry of Justice. They were the first government department who had asked to start looking at a major program of works. And one of the things that became really obvious as we looked at it is whilst the Ministry of Justice develops lots of different building types, there's a massive commonality across lots of them.

So we started to articulate this commonality to them, and start to say that's something that government could start to leverage. We said, if you were going to do this, then there's a whole number of aspects that it starts to touch upon. So you don't just design the platform components.

It changes the way you buy them, and the digital way that you start to classify things, and the kind of data structure that sits behind these. Obviously, it impacts on the way you manufacture them and the scale of that. And that then changes how you deliver things on site.

So if you think of it as a kit of parts, you can imagine it touches every single aspect of design procurement, manufacturing, assembly. And I think that was part of the mental hurdle we had to get over. We're sort of changing the mindset. It's quite different in a kind of product mindset versus project mindset.

One of the things that we did to kind of articulate why you would do this was a thought exercise with all of the major spend government departments that you can see on the right. So we plotted out a series of the key spaces that they buy in their assets, and said, everything lives on that graph if you look at clear span and clear height.

So again, thinking about the individual spaces that make up assets, if you start to plot them, you see clusters of things which tells you something about the commonality. So we said, you can imagine having a sort of kit of parts that does you the big empty shed-- so sports halls and warehouses and things, a kit of parts that does you cellular repeatable accommodation, so things like prison cells for the Ministry of Justice, single living accommodation for the military, student accommodation.

And in the middle, there's a sort of broad range of space types which are actually very common in their functionality. The reason is that, for instance, floor to floor heights are set by the height of a person plus some clear height plus the zone for structure and the kind of electrical systems. And at that sort of floor to floor height, you can get natural daylight about 8 meters into a building.

So that's why hospitals and apartment buildings and schools have about an 8 meter span. It has nothing to do with sector. It's everything to do with people. So those were the sorts of things we were saying, look, if you look for those commonalities around people, for instance, that opens up these possibilities of how you might use the same kit of parts across all of those asset types.

We also had some data from one of the government departments. They've been analyzing their construction framework, and found that of every pound they spend-- or dollar-- about half of it ended up in the actual building. That's the 51.3%.

Almost half of the money they spent went into things that weren't the building. So some of it is, obviously, very useful things. It's supply chain profit, for instance. But a lot of it is the transactional friction-- the waste, the rework, the overhead and overhead, the management of management of management processes.

So a lot of it is kind of not value adding aspects. So you said, you could start to think of this as rather than trying to squeeze a few percent out of your supply chain, you should look at the left hand side and say, well, how little material can I possibly use to make the assets? On the right hand side, if you said taking that material, what is the least number of steps, the fewest number of people, the least number of processes, and say, the highest level of automation I could use to turn raw material into a building that would inevitably be cheaper, quicker, lower carbon, et cetera?

So we took the categories you see here and arranged them on the right and said, here's how you should prioritize things. So firstly, optimize the products. Protect your supply chain profit, so make sure everyone on the supply chain gets paid well, because that will enable all of these better behaviors.

Spend less on design fees, so don't keep inventing every building from scratch. Start to change the design process to configure the components, and designers spending more time on aspect and prospect and massing and public realm and the kind of more value adding things. Reduce the overheads and overheads by being more directly engaged in the supply chain.

Eliminate the risk through virtual prototyping, physical prototyping, and again, using systems which are manufactured and not built on site. And that's how you start to change the cost and the carbon of delivering assets. So these were the benefits that we started to articulate through that process-- so less material as I talked about, trying to use fewer components, because that's how you get into that kind of manufacturing mindset.

And also, standardize processes in terms of manufacture and assembly. Because if you do that, then you can start to engage a much more diverse supply chain, for instance. So one of the big problems facing construction is the current skills gap, the aging demographic, the poor productivity.

If you could suddenly take, like IKEA, load of people who didn't know how to make furniture and turn them into furniture assemblers, that transforms the way you think about furniture. It should transform the way you think about building. So if we could use these standardized components, suddenly our labor force becomes much, much more diverse.

We tackle skills gap. We tackle productivity. And suddenly, you've addressed some of the really big challenges facing construction.

So we started thinking broadly in terms of these components. You'll see superstructure sits in the middle of this diagram, not because it's the biggest part of the cost plan of any project by any means, but it's the enabler of other things. So for instance, you can already buy a millimeter perfect unitized cladding system, for instance.

But what we found is the supply chain spends most of its time filling in the gaps between its millimeter perfect manufactured system and the traditional piece of construction it attaches to. So facade supply spend a lot of time bracketing and mass stacking and shimming and filling in the gaps to make the thing work.

What we found is if we said to them, we can control the tolerances. And we'll give you a super accurate carrier frame. What would that do to your cost? And it dramatically reduces the cost, whilst improving the quality and making the building airtight, et cetera. Similarly, mechanical electrical systems are very often prefabricated now.

So again, if you have a very accurately controlled carrier frame, it just enables that supply chain to do more and better work. And we're increasingly trying to get fit out and MEP to come as a single product, so it has not only the mechanical electrical systems, but the architectural finishes, the fire stopping. So again, you get a product that does more things, clips into a sort of known and highly controlled superstructure. And that's how you start unlocking benefits.

The bits that surround that diagram are the kind of supporting elements. So top left, one of the things that we said is really important is to start to really rationalize design-- so get data from different programs, understand where that commonality is, turn that into a set of rules that will drive these components. If you had that, you could actually codify those rules.

And you could create digital configurator, so you could automate the design process much, much quicker. And if you had both of those, you could then enable a market through a sort of digital marketplace, and a consistency of requirements that allows the supply chain to address them.

So we said, you need all of those pieces of the puzzle. But I'm going to talk a bit more about superstructure later, because as you can see, that for us is the-- that carrier frame is the bit that really unlocks a load of other things in different aspects.

I'm going to skip a few documents. That first document gave government, I think, something to think about. It was influential in what followed. So the next big step was end of 2017. In our federal budget, the government announced it was going to use its spending power to improve cost effectiveness, productivity, timeliness, through driving adoption of modern methods of construction.

It didn't, at that point, say what modern methods of construction meant. It was a deliberately sort of wide term that includes digital as well as physical aspects. But it was saying look, this is coming. We're going to start to get more and more detailed around it.

That document was then picked up by the Infrastructure Projects Authority who published, end of 2018, a very good articulation of the idea of platforms. And you can see the text there, digitally designed components across multiple types of assets. That was launched the same week the Construction Innovation Hub was set up.

So that was there to take that thinking, take the work that Ministry of Justice had done, sort of test it thoroughly, finish off the design and then propagate it out across the industry. So the Hub has been there since 2018 to really drive this thinking. It brings together digital partners, manufacturing partners, testing partners, to really start and say look, what is this kit of parts?

What are the enablers we would need to create a marketplace for these? And what are some of the things that a platform would need to accomplish? And as I said, I'm the Design Lead for that. So I've been helping the Hub in sort of propagating this thinking.

The next three documents I'm going to talk about in a bit more details. The first of them was this Construction Playbook. So this was published December, 2020. And you can see here again, it's got this reference to product platforms.

So when you have a look at this document, on the left hand side that you can see these are 14 policies which government and industry have co-developed. They're the sort of rule book for how government plans to develop social infrastructure, moving forwards. And you can see here, it's the rules which all government departments either have to comply with or explain why not.

So if you have a very particular requirement because of your in a sort of very specialist sector, maybe some of these rules aren't applicable. So if you're in a sector where there isn't an enormous pipeline, then some of the pipeline rules don't apply, for instance. But the thinking is, you have to say to government I will comply, or you have to get a dispensation.

So that's quite interesting in itself, that governments now set out a very strong framework that applies across all its social infrastructure spend. So that again, thinking moving from projects to programs to cross-sector is enormously powerful. One of the key things, key policies, is this harmonize, digitize, rationalize demands.

So just like I talked about the floor to floor sites example and therefore the structural spans, that's what that policy is getting at. It's saying, look, we're going to look across all of the government sector. We're going to start to look for these commonalities.

And we're going to start to converge departments, so we have a smaller number of more repeatable standards. That in itself would allow the supply chain to get greater certainty and to start to gear up for investment. But it's then accelerated by the piece on the right hand side saying, and we're going to use that to develop these product platforms.

So this came out end of 2020. And it was the kind of next big step in government thinking, and articulating a very, very clear approach, building on that MMC autumn statement saying, right, this is what we really mean. It's product platforms using commonality. Now, we can start to gear up to do this.

Very soon after that, the Construction Innovation Hub published a piece of work, called Defining the Needs, which I was heavily involved in. This was taking pipeline data for these departments-- so the big sort of direct procuring departments-- taking all of their pipeline data in as much detail as we could get it, applying Uniclass to it-- which was an interesting exercise, but it never been done before.

So you can imagine, every department has its own nomenclature. Every department has its own way of publishing its pipeline. Getting it into a common format, getting it all into Uniclass and being able to do these sorts of visualizations where we could cross compare departments was enormously helpful. But it also helped to really define where a product platform might start to play.

So these are visualizations of the relative scale of the pipelines we looked at. You can see vast amount of housing. That's not the complete amount of housing the UK needs. That's purely focused on local authorities, so potentially socially funded housing.

So it's only looking at small bit of the market, but you can see the amount of it is vast. We also look to education, health care, certain aspects of Ministry of Defense and the Ministry of Justice to get that kind of data set and say right, this is kind of the 50 billion pounds pipeline we're looking at over the next few years.

We did these sorts of analyses. So having talked about grids and things, we were looking at how common are spaces and how big are they? So you can see things like there's loads and loads of toilets, but they're relatively small, which might tell you something about the kind of forms of industrialized construction that you might apply, whether it's pods and things.

We started looking for commonality of spans and found that there was a huge number of-- or huge proportion was serviced by that kind of 8 meter span. So we did lots and lots of these data visualizations to look for patterns, and look for where you might start to apply some of these platform approaches.

These are some of the key findings. So I talked earlier about that kind of scatter graph. We had a much, much more detailed and realistic version of that, but it proved the same point. About 35 billion pounds or 70% of that pipeline we looked at could be serviced by an 8 meter span platform.

So that's the education, apartments, schools for the reasons I've talked about. But we also found some other really interesting things. So for instance, less than half of a health care facility is actually clinical space. There are obviously lots of very important clinical spaces, but most of the hospital is circulation and waiting and storage and plant space, and all these kind of non-clinical supporting functions.

Same in schools, we found that about 50% of it less is actually teaching space. Again, lots of it is the gymnasia. It's the dining halls. It's the circulation. It's the storage. It's all of those other supporting spaces.

So it starts [INAUDIBLE] something about the way you might let departments focus on their really clinical or critical spaces, and really focus on those. The other generic spaces you might take offline and say, maybe we start to develop them centrally. So rather than every department come up with its own solution for corridors, we could have a kind of suite of corridor types which are multi-function that we buy at scale. And again, that's how you start to get into these bigger numbers.

One of the other interesting findings I thought was that there were 104 different names for toilets across the estate. And they were similar names. They were sort of staff, hyphen wc, staff underscore wc, wc hyphen staff, et cetera. So they were similar that you could tell that they meant the same thing, but they weren't machine readably be the same.

And I think there's lots of latent standardization that's hidden by that kind of nomenclature. So by unpicking that and saying, well actually, we can tell you exactly how many toilets there are, where they are, the spaces and things, suddenly that elevates thing which is very, very common but hidden by that kind of different naming convention.

So these are some of the kind of big findings that came away from it. But it demonstrated the potential addressable market for platforms, which is vast, and the particular addressable market for a superstructure system, which is 70%. Which we sort of intuitively thought was true, but now we have the data to properly prove it.

The last of these-- I mean, you've just seen the video-- but this document came out middle of September, 2021. And it was the next big articulation of platforms. So you've seen, the video summarizes this slide.

You can see here, again, some of the vision things they're talking about. And one of the strong links, I think, that came out of this was the strong link between physical and digital, to saying if you have these, then firstly, you have a data set that underpins it. If you have that, it would enable configurators.

And so you start to not only enable that kind of manufacturing productivity, but you also allow better digital tools to start to be used-- automation, computational design, configurators. That starts to become very interesting. And I think the really key thing for me is they've now said, this is a mandated approach.

It's government said, look, we've got two years to work out what the requirements of the mandate look like. But we're going to start to mandate this. We're going to definitely use this to drive social infrastructure.

So the last time UK government did that was the BIM mandate. And that had a dramatic impact. And it obviously set our whole industry into digitalization. This is the next one of these.

You say, now you've got the hang of that. The next one is, you're moving into platform. So I think that, again, is an incredibly powerful market driver, because it's the way that markets are going to start to really respond to this when government's saying, if you want to play in the public sector, platforms is the way forward. So that, I think, is going to have a big impact the next couple of years.

I've mentioned it a couple of times. I'll just show you very quickly. One of the things we've also been working on here is new digital tools that are underpinned by these approaches.

So say if you have that kind of kit of parts, you could also generate the rules. And you could also generate the better digital ways of working. So we've been developing a number of these for public and private sector clients, over the last few years.

I'll show you the schools one, because it's probably the quickest and easiest to demonstrate the power of this. But we did a piece of work for Department for Education. We took all of their standards and their space requirements.

We turned that into a digital library of components. And we turned that into an app. This is a series of nine-year-olds in a school in North East London, who are now able to design a code-compliant school in about 15 minutes.

So you plug in what type of school it is. It gives you a digital sort of LEGO bag of parts. You pick your site and take all the buildings off. And you then cluster-- you then place all the clusters-- where you want them.

There's rules built in that mean you can't put say, a classroom over a sports hall. It has the adjacencies built in, so certain spaces only go next to each other. But you can literally design a completely code-compliant school in 15, 20 minutes.

And when you've got a school you want, you can export that into Revit or whatever your authoring software is, and off you go. And it's taken what would have been a procurement process and a design process, and it's turned it into a thing that literally, nine-year-olds who play Minecraft like, can instantly pick up and go, yeah, I can design your schools.

We're also developing these in housing. I think health care will be a great sector to do this in. But you can start to see now, this is totally underpinned by that kind of rationalized approach.

So the last couple of minutes, I'm going to show you what this looks like in practice. This is a building we've been doing for a client called Landsec. They're the biggest private developer in the UK. They're actually using what you call a Platform of 3.

So these numbers we use to describe, as you've heard, the kind of different performance characteristics. Platform 2 is the mid span one that has the sort of 35 billion pound market. Platform 3 is the next step up, so it does a 9 by 9 or 9 by 12 meter grid, which is very common in UK offices. And it's a kit of parts developed around that.

It's worth saying that as whenever we're developing these platforms, a lot of the-- a bit like the IKEA book and the IKEA Allen key or hex key-- the processes and the components are very often common across multiple platforms. So soon as we've learnt something in commercial offices, for instance, we can instantly apply it to education or health care.

So what you're seeing here is there's a whole series of some of them physical components, some of them process and ways of working that we're developing in one platform, and then instantly updating the rules and using in others, which means we're moving very, very quickly. These things are evolving quite quickly. And we're able to instantly cross-fertilize best practice from one sector into another, which is another massive hurdle we're able to cross over.

So this is a snapshot of [INAUDIBLE] CGI. The project is on site at the moment. It's two nine-story office buildings. This is just south of the River Thames, near St Paul's and the Tate, for anyone who knows London. You can see here, architecturally, not constrained by the fact it's using a kit of parts.

So you don't look at it and go, yeah, it looks a bit-- it looks like it was made out of these building blocks. Architecturally, it very well responds to the site and the planning context, but getting all of the benefits that we've been talking about from these kits of parts.

So we developed the superstructure components, very limited number of these. They get used again and again and again. It's the same beam type everywhere, for instance, which means we can then start to get into automated CNC laser cuttings.

So some of the components are laser cut. All the temporary works are laser cut and robotically welded. So they come to site with millimeter or sub-millimeter accuracy, which means that tolerances are-- we're plus or minus 5, compared to plus or minus 30 or 50 you might get on a traditional construction site.

Again, because we have that digital kit of parts, we were able to automate the design process. So we used a workflow that generates points and lines representing each of the platform components. From that, it generates a data set. And from that, we can generate the Revit model.

So the BIM model is actually a view of the data set, not the other way around, which is quite interesting. So on this project, we modeled the cores manually, but all of the superstructure-- using the platforms-- was generated automatically. So something changed in the project, we didn't update the model.

We just generated a new one, because it was quicker to do that. But then, you get that total data set that drives the procurement, drives logistics, drives everything downstream. So again, we're using this to massively speed up the generation of information, the quality of that information.

From that, we could then automate the manufacturing process. These are some of the structural components, the temporary works, that are being robotically cut and robotically welded, so super high accuracy, again, driven directly from the model, sub-millimeter accurate and completely consistent. So again, we get that kind of total consistency of a BIM approach.

When we get on to site, this was actually a prototype we built before we tried it on site. But you can see here, we're measuring things in minutes. We're using a little bit of automation and a handful of people, of color-coded components which are designed to only go in the right way.

And this was looking at whether we could build the entire superstructure without ever working at height. So you can see here, a few people with a little bit of automation lift the shuttering up from underneath, lift the brackets up from underneath, pump the concrete from underneath. It's low carbon concrete. And a little robot levels it itself, compacting.

And what we're trying to demonstrate here is that you can build a superstructure twice as quickly with half the number of people, to millimeter accuracy and with all of the fixing points. The mechanical system's already cast in, so the MEP then clips into position.

We're using in situ concrete, which I think people think is counter-intuitive, because they think offsite is generally the way to go. This is the kind of series of steps that we take to get the platform superstructure concrete in place. And you can see here, it's very, very high quality.

It has the MEP fixings. If you were to precast it, there's then a series of additional steps that are required. So you actually add cost and carbon and labor and things. So again, going back to those drivers of the least amount of material handled the fewest number of times, if we can turn the factory into effectively a manufacturing facility, then we get all of those benefits.

And we go back to those value drivers of not double handling material things. So that's why we're in situ, pouring the concrete. These are some of the benefits. You can see here, because it's very optimized, we've got fewer vehicle movements, fewer components.

The weight of the building's come down. We've actually managed to prove, independently verified, 22% less embodied carbon in the superstructure. And this is the UK's first net zero commercial development. So we're instantly getting these carbon benefits.

Facade then is a sort of kit of parts which gets clipped into position. We got a fantastic deal on the superstructure, because of that carrier frame. MEP, similarly, we've got a few types of MEPs. So Landsec can decide quite late in the day whether they want fan coil units or chilled beams.

That gives them greater adaptability in talking to potential tenants. Obviously, it gives them greater maintainability moving forward, because you can easily take out and place things. And again, the MEP's designed as a series of cassettes which interface with the platform structure, inside the structural zone.

So for quite low floor to ceiling-- floor to floor height, to about 3 and 1/2 meters floor to floor heights, we still get very good floor to ceiling heights. So we get a much better volumetric net to gross. And that leads on to a whole load of benefits.

Because we're compressing the surfaces and floor to floor heights, there's less air to heat and treat, which means the air handling units are smaller, which means the running cost is lower. So we get what we would describe as a chain reaction of benefit, where having done all these things, we're getting more and more reduction in carbon in the kind of building phase, but also lower whole life cost. And that's some of the ways we've managed to get this to be a net zero carbon building.

Last slide for me, so you can see here, these are some of the metrics we're getting from this project-- so more net lettable area, less cost, less embodied carbon. It's quicker, fewer operatives. So if you imagine taking the scale of these benefits and you multiply it by the 50 billion, you know, at least spend in the public sector, you can imagine this is how we're going to start to get some really dramatic inroads into cost of carbon and construction.

And again, the fact that it's an idea that was developed for the public sector, has been picked up by the private sector, and that's feeding back into the public sector and the way we deliver schools and hospitals, departments. I think that's a fantastic example of where the industry needs to go and how we're going to have to collaborate our way out of some of these issues.

So that's enough for me. I'll be more than happy to take some questions. And hopefully, that's been of interest.

They'll obviously be more links in the class materials. Some of the documents you can read at your leisure. But yeah, hopefully, that's been interesting, and looking forward to answering questions. Thank you.