黑洞社区

In the early nineties, Paul Cross, one of Arup鈥檚 clever men, built a spreadsheet tool for the Commerzbank Tower project in Frankfurt. He called it the Optimiser, and he used it to make sure that the 18,000 tonnes of steel in the tower were all working at more than 90% of their capacity. In modern parlance, he ensured there was less than 10% wastage. The financially savvy client immediately tried to buy the Optimiser for 1.5m marks.

Today, Clement Thirion, one of our own bright young things, is doing an engineering doctorate at UCL on the subject of 鈥減utting the material in the right place鈥�. Its aim is broadly the same as Paul鈥檚 spreadsheet 20 years ago, but Clement is looking much wider. His research began with some simple case studies, looking at the material efficiency of typical buildings. The results are bewildering - and are the tip of the iceberg as far as wastefulness in the industry goes. To take one example: in a four-storey concrete frame building, less than 40% of the concrete in the columns are actually working. The received wisdom behind this is that it is 鈥渃heaper to make all the columns the same size鈥�.

In the same case-study building, Clement found that 20% of the concrete in the floor slabs could be removed and they would still comply with the code. Much more interesting, and way beyond the code, only 50% of the remaining concrete is actually useful 鈥� the rest is in a tension zone where it does diddly squat except add to the weight of the building and boost the load on the columns and foundations, so they also have to be bigger than necessary. All in all, only about 40% of the concrete in the building is actually working. The rest is just ballast (and that鈥檚 a polite way of putting it).

Our rules reflect the value system of a bygone age. We wouldn鈥檛 watch a television designed with the technology of the fifties, so why design a building that way?

So here鈥檚 a simple suggestion - a sort of fight the flab for engineers. From now on you should only get 黑洞社区 Regs approval if you demonstrate that all of your structure is acting at, say, 90% of its capacity. Prove that your structure sits in the Goldilocks zone where it is just strong enough. Overdesign would become a professional offence. Overnight, we would move to a culture that would help those apprehensive engineers who are not sure exactly what they are doing, so they stuff in more concrete, or more steel, or more safety factors, to make sure their thing won鈥檛 fall down.

I鈥檇 go further. We have another student doing a UCL engineering doctorate; they are trying to answer the question 鈥淲hat鈥檚 the point of serviceability?鈥� 鈥淪erviceability鈥� is just a name for the set of rules that prescribe how buildings should behave in use. It runs through every aspect of building design. This student is challenging received wisdom on, say, how far a building may move. For example, many codes require that, under load, the middle of a floor should sag by less than, say, 30mm over a 10m span. On facade lines, this requirement is often stricter, say, no more 20mm over 10m, usually driven by an aesthetic desire to keep the joints between the facade panels as thin as possible. In lift and escalator shafts, we鈥檝e often been given movement limits of no more than 1 in 1,000. To hit all of these often arbitrary limits, engineers add even more material to something that is already too strong. I, for one, cannot even see a floor sagging by 30mm over 10m. Anyway, it would never sag that much because it will never encounter its over-conservative design load. And although I love beauty and refinement of detail as much as the next person, does it really matter if the joints on the skin of a skyscraper are 50mm wide when most of the time they are looked at from half a mile away?

Here鈥檚 a simple suggestion - a sort of fight the flab for engineers. you should only get 黑洞社区 Regs approval if you demonstrate that all of your structure is acting at, say, 90% of its capacity

Rules like these reflect the value system of a bygone age and bygone technologies and people who persist in using them need to bring themselves up to date. They wouldn鈥檛 watch a television designed with the technology of the fifties, so why design a building that way?

There are many culturally accepted homilies behind these anomalies 鈥� for example, you can鈥檛 let the floors sag too much because they will 鈥渃rush鈥� the partitions; it鈥檚 鈥渃heaper鈥� to use standard components; 鈥渓abour is more expensive than materials鈥�. None of these stand up to even a moment鈥檚 scrutiny. Isn鈥檛 it about time we stopped turning a blind eye to culture of overdesign that, in real terms, brings us absolutely no benefit? Instead, roll on intelligent action on the way we use our limited supply of materials. More power to Paul and Clement.

Chris Wise is director of Expedition Engineering