Image: 13459830 © Joseph Gough | Dreamstime.com

Analysis

Structural steel’s BIM challenge

29 March 2021

Kevin Campbell is chair of the British Constructional Steelwork Association’s (BCSA) Digital Technology Group, which started out as the Tekla users’ group (Tekla being the Trimble 3D CAD software that is the dominant player in the steel sector) until changing its name five or so years ago. He is also MD of Severfield subsidiary, CMF.

Tell us more about the group.

The steel sector has a relaunched digital technology group. Tell us more.
The steel sector’s position was ‘we've been doing BIM for years, waiting for the rest of the world to catch up’, but of course the world does catch up.
The British Constructional Steelwork Association doesn’t want the sector to fall behind. So we renamed the old Tekla users’ group to digital technology group. Now, as well as reviewing the latest updates to Tekla, we look at the wider potential for digital technologies, whether in a factory or on site or in the office: cloud point scanning, Hololens, integration into other software packages.
Steel is in a good position. There’s a view that we’ve been doing 3D modelling for many years and that’s correct. But we’re not sure of what’s coming next within BIM – what will digital assets of a building will look like.
Is product data being commonly used by steel fabricators? 
When we create a 3D model of a steel building in Tekla, we fully detail it and generate loads of information that we use for ordering steel, generating drawings, exporting into manufacturing and management software packages. And, of course, there’s loads of information that gets generated as part of the production process and the erection process in terms of material test certificates, welder approvals, inspection records, paint records, and so on. 
Tekla’s view is this is not going to fit in the model: you can’t export that to an IFC and back into Revit because it’s a potentially massive amount of data and a model is not created as a data warehouse.
So there’s work to be done if we want to be able to create a data pool for a structure, whereby you could open a package, and interrogate a model and click on a piece of steel or a piece of wiring to see its life history and where that part or that component was manufactured or processed, who assembled it.
But what does the end-user, the client want? Would anybody ever use that level of detailed information?
Depending on the contract, and the project and the purpose of the structure – if it's an oil refinery or processing plant - then working to the actual laid down specification can be onerous. A client on occasion may come in and want to see the material test certificates or welder approvals and audit the fabricator to make sure that everything's in line. But in reality, on most projects, does the client appoint a third party to come and check? No.
Is steel looking at robotics?
Robotic assembly is great if you’re manufacturing something small. The steel beams we handle could be up to 20m long, and could weigh 10t, so the handling of the material is the biggest challenge.
We looked at the application of robotics at Severfield and whether it was viable. We felt the technology wasn’t quite ready. While the [robotic assembly] systems are quite flexible, there are only certain types of products you can give them to fabricate and weld because of its limitations. So
you can only use robotics for 40% of your work; the rest has got to be fabricated or welded on conventional lines.
Virtually everybody in the sector builds to order for just-in-time delivery: we don’t build by product families. There will always be the need to do things conventionally due to the range of components the sector produces.
There are maybe two or three steel fabricators that have taken the plunge, buying robotic assembly and welding systems. In the UK, it’s fairly limited, because it’s pretty expensive. 
What other digital technology is being used in the steel sector?
The challenge is making tech like barcoding [of steel products] and HoloLens mixed reality link with our production systems and Tekla.
We'd like to be able to put on a HoloLens in the factory, call up a virtual image 3D image of a part, look at the actual piece of steel in front of us, superimpose one over the other, and then check it. We found the resolution and different aspects of it weren't quite good enough for that particular application.
It’s got a limited application for some of the more complex stuff, but have the ability to show consulting engineers and architects images of a structure anywhere in the parts of the world can’t be ignored.
If you’re building steel for conventional warehouses, it’s probably not worth investing in [that type of technology], but if you're building for an oil or a gas project, I’m sure there would be lots of applications.
Are digital skills an issue?
In the structural steel sector, digital skills are there in pockets. Most fabricators don't have the resources to employ people full-time in digital roles. This is a challenge. 
The bigger businesses have got resources for digital research and development, whether it’s just handling information easier, in terms of processing materials and production, as well as trying to use that information to automate stuff that's done manually, whether it's done in the office, or in the factory

The steel sector’s position was ‘we've been doing BIM for years, waiting for the rest of the world to catch up’, but of course the world does catch up.

The British Constructional Steelwork Association doesn’t want the sector to fall behind. So we renamed the old Tekla users’ group to digital technology group. Now, as well as reviewing the latest updates to Tekla, we look at the wider potential for digital technologies, whether in a factory or on site or in the office: cloud point scanning, HoloLens, integration into other software packages.

Steel is in a good position. There’s a view that we’ve been doing 3D modelling for many years and that’s correct. But we’re not sure of what’s coming next within BIM – what digital assets of a building will look like.

Steel is in a good position. There’s a view that we’ve been doing 3D modelling for many years and that’s correct.– Kevin Campbell, BCSA

Is product data being commonly used by steel fabricators? 

When we create a 3D model of a steel building in Tekla, we fully detail it and generate loads of information that we use for ordering steel, generating drawings, exporting into manufacturing and management software packages. And, of course, there’s loads of information that gets generated as part of the production process and the erection process in terms of material test certificates, welder approvals, inspection records, paint records, and so on. 

Tekla’s view is this is not going to fit in the model: you can’t export that to an IFC and back into Revit because it’s a potentially massive amount of data and a model is not created as a data warehouse.

So there’s work to be done if we want to be able to create a data pool for a structure, whereby you could open a package, and interrogate a model and click on a piece of steel or a piece of wiring to see its life history and where that part or that component was manufactured or processed, who assembled it.

But what does the end-user, the client want? Would anybody ever use that level of detailed information?

Depending on the contract, and the project and the purpose of the structure – if it's an oil refinery or processing plant - then working to the actual laid down specification can be onerous. A client on occasion may come in and want to see the material test certificates or welder approvals and audit the fabricator to make sure that everything's in line. But in reality, on most projects, does the client appoint a third party to come and check? No.

Is the steel sector looking at robotics?

Robotic assembly is great if you’re manufacturing something small. The steel beams we handle could be up to 20m long, and could weigh 10t, so the handling of the material is the biggest challenge.

We looked at the application of robotics at Severfield and whether it was viable. We felt the technology wasn’t quite ready. While the [robotic assembly] systems are quite flexible, there are only certain types of products you can give them to fabricate and weld because of its limitations. So you can only use robotics for 40% of your work; the rest has got to be fabricated or welded on conventional lines.

Virtually everybody in the sector builds to order for just-in-time delivery: we don’t build by product families. There will always be the need to do things conventionally due to the range of components the sector produces.

There are maybe two or three steel fabricators that have taken the plunge, buying robotic assembly and welding systems. In the UK, it’s fairly limited, because it’s pretty expensive. 

What other digital technology is being used in the steel sector?

The challenge is making tech like barcoding [of steel products] and HoloLens mixed reality link with our production systems and Tekla.

We'd like to be able to put on a HoloLens in the factory, call up a virtual 3D image of a part, look at the actual piece of steel in front of us, superimpose one over the other, and then check it. We found the resolution and different aspects of it weren't quite good enough for that particular application.

It’s got a limited application for some of the more complex stuff, but the fact that it has the ability to show consulting engineers and architects images of a structure anywhere in the world can’t be ignored.

If you’re building steel for conventional warehouses, it’s probably not worth investing in [that type of technology], but if you're building for an oil or a gas project, I’m sure there would be lots of applications.

Are digital skills an issue in steel?

In the structural steel sector, digital skills are there in pockets. Most fabricators don't have the resources to employ people full-time in digital roles. This is a challenge.

The bigger businesses have got resources for digital research and development, whether it’s just handling information easier, in terms of processing materials and production, as well as trying to use that information to automate stuff that's done manually, whether it's done in the office, or in the factory.

Main image: 13459830 © Joseph Gough | Dreamstime.com