Three men and a prefab

It’s gone pretty quiet on the prefab front, but three people working with modern methods of construction – Stuart Piercy, Andrew Matthews and Elliot Lipton – believe it’s just a case of adapting the methodology to present needs.

Words Hugh Pearman | Portraits James Bolton

Whatever happened to prefab? By which I mean, where have we got to with modern methods of construction, or MMC?  In its volumetric form (whole-room pods) it was quite a thing from the mid 1990s to the mid 2000s.

In London, Dickon Robinson, then of the Peabody Trust, delivered buildings such as Murray Grove by Cartwright Pickard and Raines Court by AHMM. In Manchester, Urban Splash dabbled with volumetric in its MoHo housing scheme in Castlefield, designed by ShedKM. In Birmingham and Leeds, the Joseph Rowntree Foundation built its Caspar blocks (City Apartments for Single People at Affordable Rents) – Birmingham by AHMM, Leeds by Levitt Bernstein, which used volumetric and semi-volumetric techniques. There was one notable failure, the Leeds Caspar, which despite winning several awards, succumbed to construction defects, was deemed potentially unstable in high winds, and was evacuated. This was a design and build process and the architects were exonerated.

Perhaps the Leeds affair spooked the market, because we heard much less of the volumetric side of things for a while after that – even Cartwright Pickard confined itself to prefabricated timber cladding sections by the time it built its well-received Bourbon Lane housing development in London’s White City in 2008, while BDP’s two ‘Abito’ micro-flat developments in Manchester and Salford used in-situ tunnel-form concrete ‘cells’ into which were craned kitchen/bathroom pods containing 95% of each flat’s services.  Developer First Base’s multiple award-winning Adelaide Wharf housing development in east London by AHMM used fast-track office-building systems of framed construction with a unitised cladding system – no scaffolding required.

MMC, then, is about a lot more than volumetric construction. And in equally good health is engineered timber – essentially, solid-wall timber prefab construction of various kinds, as most spectacularly demonstrated by Waugh Thistleton’s nine-storey Stadthaus in Hackney (at Murray Grove again), the tallest of its kind in the world (RIBAJ, March 2009). So where do we go from here?

In search of answers, we assembled three people in the know: architects Andrew Matthews of Proctor and Matthews and Stuart Piercy from Piercy Conner, with developer Elliot Lipton of First Base. Proctor and Matthews first got into the prefab game when working on the Greenwich Millennium Village – using ‘steel stick’ prefab wall frames made and riveted together on site. Now they are rolling out the ‘SLO’ development of 77 modular-construction two-storey houses in Harlow, using the FutureForm system of steel-framed room pods made in Wellingborough in Northamptonshire. For Matthews, the drive was not cost but quality.

‘What we tried to do in Harlow was create a courtyard house of the same area as a standard volume-housebuilder cellular box, but with a greater perimeter wall area and a better relationship between inside and outside – bringing light into circulation spaces. We’re using prefabrication not for efficiency, but to make a better product. 

‘We’ve learned a lot up there. One of the issues with prefabrication in this country is the way we procure domestic architecture, which doesn’t lend itself to the way prefabrication is set up. Most prefabricators are looking through the telescope at efficiencies and cost savings – which requires lots of unit numbers running through a factory – but we don’t have a housebuilding industry that can order lots of units. Housebuilders are geared to cash flow and sales rates. They don’t want to buy modules upfront, all fitted up, because they want to roll out housing only at the rate they can sell it.’

That, according to Matthews, is looking through the wrong end of the telescope.

‘If you go to Japan, they have automated systems to make bespoke housing. They’re not necessarily dependent on volume. Toyota’s prefabricated houses are all different, but assembled from the same set of components. They target a high end market. In the UK we’re a very, very long way away from that kind of market.’

On the other hand, he says there is no market resistance to advanced construction techniques. ‘The consumers think the space and the arrangement are fantastic. They’re not concerned with what a house is built of. As long as it can be insured, has BRE approval and NHBC cover, they’re not really that worried. They’re interested in something that’s different which they can buy for the same or less money than the standard Barratt house down the road.’ 

This is just as well, given that Proctor and Matthews has turned part-developer itself to build out the Harlow scheme following the withdrawal of a housing association.

Where Proctor and Matthews’ recent work has been in the field of suburban housing, Piercy Conner has done a lot of research into the urban alternative, principally through its unbuilt ‘Microflat’ concept. It has also done plenty of more conventional work as well.  But its latest project, now on site at Rajarhat, a new town outside Kolkata in India’s West Bengal, is a demonstration block of 12 apartments based entirely on flat-pack and steel stick components. This project is one of several run by the international trade initiative Living Steel, a collaboration of steel makers including Tata Steel, which builds housing in urban centres and is managed by the World Steel Association. (Cartwright Pickard is just completing others in the programme for China’s Sichuan earthquake zone).  Steel might at first seem a curious choice for hot, humid climates – but the secret here lies in solar shading coupled with cross ventilation. Although this scheme is vertically stacked rather than single-storey, one thinks of Jean Prouvé’s late 1940s series of prefabricated ‘Maisons Tropicales’ with their sliding screens and roof vents.

‘I guess this project is a bit of a look back to my Grimshaw days,’ reflects Piercy. ‘I was working on a scheme at Zurich airport, and they were bringing in house-sized components that were assembled on the other side of the airport to reduce the level of activity on site. Here, the idea was to ­re-introduce steel to the domestic market. The way it worked was that basically, as long as the parts were in the Tata catalogue, you could use them to make the apartments. Instead of being volumetric prefabrication, it’s component-based. It’s modular in the sense of being dimensionally stable – you can take a panel off and stick another one on, or take a whole wing off and replace it with another, and the bolts will be in the same place and everything will fit.  It’s in the catalogue and off the shelf – we build the largest components we can in the factory that can be delivered flat on a lorry.’

As labour costs are low in India, says Piercy, the flatpack approach, requiring rapid site assembly, works as long as the system for putting them together is straightforward. There’s no need to construct huge volumes in the factory before transport.  But in fact, he says, this seems to be valid everywhere.

‘Our experience is that elements of prefabrication work really well, but prefabricating the whole thing doesn’t seem to work out.’

Lipton agrees. ‘Prefabrication is a much broader term than volumetric,’ he says. ‘The work Cartwright Pickard and AHMM did in proving volumetric was informative for all of us – we’ve learned from that what worked well and what didn’t. They were trying to capture a moment – our collective desire to improve the product and the construction process at the same time. Volumetric somehow emerged from that as a potentially euphoric solution to both problems. But what we’ve all collectively discovered, and played with, and proved, turns out to be more complicated than that. There are lots of ways both to improve the product and to improve the construction conundrum.’

For First Base, he says, there is a whole list of requirements that a project has to hit every time. Prefabrication is one way to achieve the end result, but only one. For instance, the volumetric approach works well for hotels – where the rooms are small and so easy to prefabricate and transport, and tightly packed, kitted out with identical services, making them easy to slot into place. Permanent housing however, with its variety of room types,  is not the same game at all. Hotels and student housing projects also have much shorter lifespans.

‘For student housing, you might expect a 25-year life – but when we design a house, we’re looking for a 200 year life. Are we clear with volumetric that it can deliver a 200 year product life? – No, we’re not clear. But maybe for a hotel or student project that doesn’t matter as much.’

Matthews agrees that the way forward may be for domestic architecture to learn from office building techniques – such as the variation in sophisticated brick-faced precast panels you now routinely see on new commercial buildings. ‘They are a work of art, and they are bespoke to that building,’ he says. ‘We can never access that technology for housing.’

Ah, but you can, says Lipton. ‘We use panellised facades on just about all our buildings’.

But the difference is that these are large, multi-storey urban apartment buildings rather than small-scale individual homes. Typically First Base will also prefabricate its bathrooms offsite, and is increasingly looking at other elements such as the electrical wiring looms, risers, and internal partitions of the development.

‘As an industry we must embrace all these elements which drive towards a better process and delivering a better product,’ he says.

For Piercy, the prefabrication movement never paused, it just developed new thinking.  ‘There’s been a quiet revolution going on,’ is how he puts it. ‘Lots of small tweaks in manufacturing have led to masses of prefabrication. At the Athlete’s Village which we’re working on for the Olympics, we’re using precast unitised whole-facade elements.’ 

It’s the kind of industry that was first tried in the 1960s – with some well-documented dire consequences – but which has finally lived up to its promise thanks to computer-aided design and manufacture.

There are some state of the art factories in the UK, from the likes of Kingspan and Laing O’Rourke, plus of course plenty of others in the rest of Europe.

But then, you could argue that this is nothing new. Look at the Victorian house building industry, once analysed by the Smithsons who were smitten by its ability to mass-produce components that could be used in combinations to produce streetscape variety in otherwise standard house types. It was also the Victorians, of course, who pioneered the flat-pack transportable building in the form of field hospitals, chapels, botanic glasshouses and so on. 

The message coming out of our discussion was clear: every generation has to find its own best form of component-driven architecture. And without any particular fanfare, no Eureka moment, we are now living at a good moment in its evolution.