Your browser is no longer supported

For the best possible experience using our website we recommend you upgrade to a newer version or another browser.

Your browser appears to have cookies disabled. For the best experience of this website, please enable cookies in your browser

We'll assume we have your consent to use cookies, for example so you won't need to log in each time you visit our site.
Learn more

Machine learning 'would have prevented Grenfell fire', Autodesk boss claims

3122255 grenfell tribute

The Grenfell Tower disaster “could never have happened” if modern technologies had been in place, according to Autodesk chief executive Andrew Anagnost.

Speaking to New Civil Engineer at AU Vegas, Anagnost said that advancements in machine learning coupled with generative design will ensure that everything built in the future will be instantly and actively monitored to ensure increased safety.

He said that technology will be able to predict and prevent catastrophes such as the Grenfell fire which killed 72 people in June 2017.

“It’s early on but machine learning will absolutely improve building and fire safety,” Anagnost said. “We can already simulate the impact of natural disasters fairly well on cities. Our biggest problem is not knowing what needs to be built but making sure it does get built the way people think it should be.”

“Most of the problem with unsafe construction isn’t that people didn’t know what needed to be built, it is that on site the wrong thing got built. Too often the wrong material is used or isn’t built to the exact specification and then you often have buildings not being as safe as people expected.

He added: “Look at the horrible high rise fire in London that has basically been traced back to the materials being used were not the ones that were supposed to be used and nobody knew. They basically built a big candle which resulted in a horrible fire.

“All of that could be avoided by machine learning algorithms that do automated checks which say material X was installed, material Y was required and then it flags it as an uncertifiable building. But nobody knew, or if they did know it got lost somewhere along the way.

“Machine learning is absolutely going to be able to identify those kind of problems so that what gets built is actually survivable under the situations that it could come up against.”

The government has pledged to introduce a blanket ban on combustible materials following the Grenfell disaster, after expert witness reports which determined the cladding directly led to the fire spreading up the building. A group of expert reports concluded that the cladding used at Grenfell Tower was never tested, did not comply with fire regulations and was incorrectly installed.

Earlier this week New Civil Engineer reported on what impact such a ban would have on tall timber buildings, with Timber Research & Development Association membership and marketing manager Rupert Scott saying a blanket ban was “unjustifiable”.

Like what you’ve read? To receive New Civil Engineer’s daily and weekly newsletters click here. 

Readers' comments (1)

  • No programmed check of a product specification will exceed the ability of a building controller. As I understand it the issue was not the lack of a check, but the interpretation of the rules.

    Generative design and machine learning are great technologies, but making safety critical judgements from ambiguous building regulations is probably not their best use.

    Andrew seems a highly trained individual, but has no experience in the construction sector.

    Unsuitable or offensive? Report this comment

Have your say

You must sign in to make a comment

Please remember that the submission of any material is governed by our Terms and Conditions and by submitting material you confirm your agreement to these Terms and Conditions. Please note comments made online may also be published in the print edition of New Civil Engineer. Links may be included in your comments but HTML is not permitted.