Why adaptivity is the key to digital twins

Let’s start with a “what”. According to Abdulmotaleb El Saddik’s [1]:

He continues his definition with a “why”:

This embodies the challenge we see with digital twins today: client focus too much on the “what” (i.e. – we’d like a digital twin) and not enough on the “why” (i.e. the benefits that they are hoping to achieve).

The first step in creating a digital twin is to create a Common Data Environment and digitise the representation of a building, piece of infrastructure or other asset. This could come from re-using the BIM model used to to design it. But if it is an older asset, this might mean point cloud scans to create an accurate (planar) representation of the asset. Or the digitisation of 2D plans into a 3D model. This can be time-consuming, difficult and expensive. However, on its own, creating a digital model is pointless as it provides no benefit to the asset owner.

The second step is to bring the digital model into a software environment where it can be displayed and manipulated, where information can be overlaid and stored, and where calculations and simulations can be undertaken. A number of software companies, such as Ecodomus and Willow, provide digital twin packages that allow an accurate visualisation of the 3D model, so that the user can go on a virtual tour of the asset. We recently saw a demonstration of a 3D model (presented as a “twin”, but we do not credit it as such) of Sydney Opera House. The user could move around the virtual building and click on various hatches and see the maintenance manual or log. But all this reminded me of – and I am showing my age here – was when I used to play computer games such as Doom. Without some benefit to the building is run, clients might as well save their money: it’s pointless.

This is why, at Priestland Consulting, we have spent time developing a library of use cases for digital twins, which are the starting point for whether any investment is worthwhile. We have grouped the use cases around a number of categories: workplace efficiency, power management, utilisation factors, maintenance reporting, occupancy levels, energy consumed by systems etc. Clearly a number of use cases can combine to create the business case for the investment in a digital twin; and you only need to invest once in the Common Data Environment and the visualisation software. But it is hard to make the use case stack up. We did some early work for a utilities client on the potential for a digital twin at a railway station. We could see the danger that they would invest £250k in a system that would save them £7 by predicting that a single light bulb would fail, and rapidly brought the project to a close.

But there is another fallacy in and around digital twins, and that is to do with the way in which we design buildings and infrastructure. At present, our design philosophy is static. To explain what this means, imagine the frustration of explaining to a building owner, who has just invested in a digital twin, that his building is 40% less efficient than it might be. And telling him that, he can get the benefits of this advice in another 23 years’ time when the building reaches the end of its design life and we can design and build him a better one. That really would be pointless.

The new world with digital twins, smart sensors and dynamic optimisation needs a new approach to design: buildings need to be adaptive, not static. As engineers, we need to design into buildings new ways in which they can adapt and change: room partitions that can move more easily; MEP systems that can alter the mood and tone of rooms; signage and wayfinding that allows people to be directed in different ways; flexible ‘plug and play’ locations for power points and connections. That way, when the digital twin shows that the building is inefficient, there are ways in which it can be improved. A learning building, in other words. And you cannot learn if you cannot change. To misquote Charles Darwin, it is not that we want to design the strongest or most intelligent buildings, but rather to design the buildings that are most adaptable to change.

It is therefore adaptivity that will allow digital twins to meet the “why” of El Saddik’s definition. Adaptive buildings will help us to monitor, understand and optimise the building’s functions, thereby improving the quality of life and well-being of users.

More News