With MMC shifting construction from the site to the factory, mass customisation offers variety. But how to manage it?
‘The singularity’ envisages a switch from mainly traditional on-site construction to mainly modern methods of construction (which go by various names including prefabrication, volumetric, modular, offsite manufacture) using design for manufacture and assembly (DfMA). In it, all that is left for the contractor to do is to prepare the ground and assemble the modules as they arrive on site.
The pre-manufactured value (PMV) of this move is well-rehearsed – safer, faster, less wasteful, better quality, more sustainable, and so on. Extracting the fullest possible benefit from it is less well understood, however, presenting designers and the supply chain with significant but exciting challenges.
Maximising PMV requires architects to value engineer their own output. It requires design to a level of detail that is capable of being manufactured. It means tessellating standardised kit-of-parts components to fit a standardised structural frame. It means early-stage planning in collaboration with the whole supply chain to optimise off-site production, delivery and on-site assembly along a purely digital workflow.
In short, it means emulating the 50-year old manufacturing revolution that has transformed, for example, the automotive industry, with the benefit of being able to learn from its mistakes. The laggard, inefficient but vital construction industry will be transformed.
Of course, there is an elephant in the room. Construction has innate physical constraints and the baggage of centuries of stubborn habit and bureaucracy standing in its way. These legacy roadblocks are real and, in many instances, important facets of the governance checks and balances that hold the built environment accountable to the stakeholders it serves.
Perhaps the most important consequence of these roadblocks has been to require that, to a greater or lesser extent, every new building or asset is a one-off prototype. Unfortunately, this fundamentally holds back the standardisation required to make the singularity happen. To bastardize Henry Ford’s alleged quip about Model Ts, offering any colour of building so long as it’s black is a non-starter.
Individualising the prototype
A newer concept is changing all that: mass customisation. It allows you to satisfy multiple different customer needs from the same production line and supply chain without unduly affecting the (low) unit cost. Also called personalisation, made- or built-to-order, the concept is already mature in many areas including – you guessed it – the automotive industry. David Miller of DMA has a growing and mixed experience of this nascent concept. ‘Mass customisation is about being intelligent about what it is that you break down into a system. Do as much offsite or in factory conditions as you can.’
Bryden Wood’s Crossrail project to design and apply linings to tunnel walls exemplifies how this might look. Starting from accurate digital scans of the tunnels, the architect’s design – a panel-and-ladder frame system using SolidWorks software – directly informed manufacture. Because one CNC machine was able to produce a variety of different moulds for the panels, it was possible to get a bespoke solution from a standard process. Working from accurate scans allowed a design to fine tolerances and minimised the overall variety of panel shapes. With the ladder frame accurately in place, the sequenced panels went in much more quickly than predicted, with no on-site workarounds needed.
The idea comes with compromises, as Jane Burry discovered when her team at the Swinburne University of Technology tried to take its bespoke FABPOD into production. ‘Disappointingly, our best viable mass customisation offer boiled down to three options: cheap, standard, and expensive. It was a real revelation to me: architecture-to-product design is not a continuous spectrum – we have different ways of looking at the world.’
Globally, the industry and its investors sense that the confluence of technology, innovation, motivation, and societal need has hit a sweet spot. Starting with housing, they now hope economies of scale will allow more customisation without affecting commercial viability. Indeed, the sector is predicted to grow rapidly, reaching an annual value of $215 bn by 2025, equivalent to around £167 bn.
Recent deals include Japanese firm Sekisui’s hook-up with Urban Splash, Goldman Sachs’ investment in TopHat, and Softbank’s gargantuan $865 million punt on Katerra. Even the UK government is getting involved, with Homes England backing Ilke Homes to the tune of £30 million.
One danger, of course, is that the fixed overheads of a manufacturing model can’t so easily flex with property cycles. Another is that you just transpose what happens on site to the factory without fundamentally rethinking how value is added. This leads to what Daniel Hall, assistant professor at ETH in Zurich calls the ‘mirror trap’. His academic research identified three ways to escape: spin-off, vertical integration, or digital systems integration. Katerra, Blokable, Project Frog and Boklok, for example, are experimenting with some of these.
Get on the platform
An important part of resolving the mass customisation paradox is the concept of platforms, a physical system with a digital twin. Bryden Wood, which has been advising the UK government about the issue, defines platforms as ‘integrated systems made from components (products or sub-assemblies manufactured by a range of suppliers) with known interfaces that can be combined in a consistent and well-defined way to create high performing assets.’ Not only was it involved in a project that produced a universal connector – SEISMIC – it has also developed a number of generic structural frameworks to suit building types needing different clear spans to underpin the system.
The notion is seen to have immense social, economic and environmental benefits, and so the newly established Construction Industry Hub is exploring the idea with the Manufacturing Technology Centre. They have invited product suppliers to develop ‘kit-of-parts pre-engineered solutions’ that combine digital, design and manufacturing principles, supported by ‘a clearly defined dataset describing attributes and performance criteria’. The example they give is ‘plug-and-play mechanical and electrical components that fit into a manufactured, insulated and finished wall panel’ which can be installed on site ‘without the need for traditional crafts and trades’.
Across the pond, ProjectFrog is leaping ahead. It has just released a piece of software called KitConnect in collaboration with Autodesk that aims to enable the digital flows necessary to bring this systematisation into being. According to vice president Mike Eggers, the system standardises how kit-of-parts information is organised so that it can be easily understood by anyone who subsequently encounters it, regardless of their role or the tools they use to discover it.
Beneficial though moving offsite is, it won’t have the required impact on productivity unless it is scaled up. In the realm of grand publicly funded social and industrial infrastructure projects, it should really offer extra value. The taxpayer will get more for less and the sector’s capacity will more likely be able to meet the major projects capital pipeline.
There will surely be teething trouble around intellectual property, standards, and rules, to say nothing of aesthetics. Nonetheless, with so much impetus behind it, as the UK government’s MMC tsar Mark Farmer put it recently: ‘The genie is out of the bottle. Change is coming.’
Implications for architects
One of the first things Farmer was keen to stress after his appointment as the MMC tsar in November 2019 was that architects should learn much more about MMC systems and DfMA. ‘The design profession doesn’t need to see it as a threat if it thinks intelligently about how it interfaces with technology and manufacturing principles,’ he said.
David Miller is clear that architects should influence what the mass customisation options are. ‘If we don’t engage, the system will produce ugly boxes. If we do, it could be quite special.’
Emma Hooper, information specialist with Bond Bryan Digital, thinks early collaboration is the root of success. ‘It’s not separate disciplines any more, but people with different bits of knowledge that you need to piece together at the right time.’
Cambridge University’s Dr Michael Ramage, director of the Centre for Natural Materials and academic advisor to the CDBB, wants architects to reframe how they think about design. ‘Work out what you can do for a construction factory that is different to a site. Ask yourself, “If I never printed a drawing again, how would I get my building built?”.’
Andrew Anagnost, CEO with Autodesk, is convinced that more and more buildings will be fabricated off-site. Success will come if you have to have ‘tight quality control, model-based processes, tight flow control, and good logistics planning; you start to look like a manufacturing company,’ he says.
See more on the digital revolution:
Is The Singularity tech’s new dawn?