This feature originally appeared in issue 109 of Position magazine.
Digital twins are touted as the business transformation tool of the next decade. Can a BIM model on steroids really do that?
If you’ve been following geospatial trends in a cursory manner over the past few years, the term ‘digital twin’ might have already worn out its welcome. The phrase seems suddenly everywhere – you might be pardoned for wondering what the hype’s all about. Is it just another corporate buzzword? Could the childlike glee and brazen claims of industry leaders all be an act?
If any of the above sounds familiar, you’ve not been paying enough attention – and you certainly haven’t been hands-on with an honest-to-goodness digital twin. Market Research Future has projected the global market for digital twins to grow from USD 2.15 billion in 2017 to around USD 35.4 billion by the end of 2025 – more than a 16-fold expansion in less than a decade.
Locally, the ground is being prepared for dominance of the digital twin, too. Key spatial peak bodies ANZLIC, FrontierSI and the Smart Cities Council of Australia and New Zealand have been hard at work on initiatives with a range of government and industry actors that should see unprecedented capability to facilitate digital twin projects in the near future.
Digital twins evade a precise definition at this stage of their development and proliferation – you’ll get a different answer depending on who you ask. This is in part because they are a maturing technology, but also because they can take vastly different forms depending on their use case. FrontierSI uses a definition adapted from Gartner in 2017, a ‘dynamic digital representation of a real-world object or system.’ ANZLIC refer to twins as ‘highly advanced digital representations of the real world’, making the distinction of ‘spatially enabled’ digital twins, in which models are contextualised with geospatial and environmental data.
The semantics may not be critical to understand, but the capabilities are. So let’s take a look at what sets twins apart from past spatial modelling and analysis techniques, and try to understand what all the fuss is about.
Models ain’t models
They’re just fancy BIM models, aren’t they? The view isn’t as facetious and reductive as it may sound. Digital twins include spatial modelling techniques, currently primarily used by the AEC industries, that have evolved from prior modelling disciplines. It could be said that BIM techniques are a part of the evolutionary history of digital twins – an iterative process of 3D tools for design, planning and construction that evolved from the first CAD suites. While this may reflect the generative development of the software and its capability, it doesn’t necessarily reflect the trajectory or potential of digital twin technology.
Whilst digital twins typically do contain highly accurate, metadata-rich 3D models and point clouds, there are significant distinctions to be made in terms of what twins can do, and what they can represent to an organisation when compared to even the most precise and well-annotated BIM model.
Stay up to date by getting stories like this delivered to your mailbox.
Sign up to receive our free weekly Spatial Source newsletter.
Brian Middleton, Vice President of Bentley Australia and New Zealand, refers to a static BIM model as a ‘digital snapshot’, with a key distinguishing feature of a digital twin being the ability to model the environment around it including a range of different data sources and modes.
“You can’t have a sustainable digital twin without managing time, location, change and context,” he said.
“So whilst a digital twin may include a 3D representation, it will also include the context of the environment in which it resides, and you’re managing the continual change of all data over time, including non CAD- or BIM-related data.”
For Mr. Middleton, the crucial difference lies in the ‘live’ aspect, the tightly integrated temporal relationship between the actual asset or project and its digital counterpart, which must reflect its environment and changes as-built and as maintained, not necessarily as-designed. Redesigns, sequencing by contractors with flow-on impacts to the designs affect the final placement of elements on site.
“If somebody says a digital twin is a representation of a physical asset at a point in time, what happens when there’s change?” he said.
“Suddenly your representation is no longer accurate. Is that a digital twin, a digital ‘sibling’, or is it more like Danny DeVito and Arnold Schwarzenegger?”
Mr. Middleton takes the view that the promise of BIM was never fully realised, and that viewing the technology from this angle may represent a cautionary tale – CAD managers turned BIM managers may not be the best custodians of the digital twin for an organisation or group of stakeholders.
“We haven’t fully realised the benefits from BIM, in my opinion. We’re still seeing projects that are continually failing to be delivered on time to budget, so we don’t want digital twins to become BIM mark two,” he said.
“If we think in terms of a workflow, CAD to BIM to digital twin is probably a fairly good three-step process. I believe the upside to digital twins are clear and obvious to all parties, whereas the benefits of BIM were never evenly split between design-and-construct versus operate-and-maintain – the builder and the client or owner.”
Bentley’s bread and butter is major infrastructure, and the capability of their product portfolio is expanding over time from design- and construction-focused tools to a collection of integrated suites for digitally managing an asset as complex as a power station throughout its operational life cycle – not just management of the construction project.
So from Mr. Middleton’s perspective, the final piece of the puzzle for a true digital twin is bringing everything together to facilitate such a multifaceted capability. Combining design systems data (CAD, BIM) with operations technologies such as SCADA, librarians and maintenance-management systems with the engineering data – datasheets, standards, specifications – in a wholly unified system.
“The big differentiator is that information is not sitting in a single system. It’s a federated view. So a simple way of explaining, is like Google or your business, Google for your infrastructure,” he said.
“But if you’re doing that in your own business, you don’t get 64 million returns on your search. You only get relevant returns because you’ve actually modelled the relationships between the data sources, based on what’s most valuable to you.”
Horses for courses
For others, different factors define the distinction between past digital modelling and analysis practices in planning and construction and the promise of the digital twin. Dr. Samad Sepasgozar is a researcher in construction technology at UNSW’s school of the built environment. He has been a lecturer in civil engineering, construction and architecture for ten years, and has developed a framework to understand the complex issues around the practice of technology adoption within his research activities. He’s currently developing a digital twin for an excavator that can be controlled through its digital twin – which shapes his view of a digital twin’s essential components.
His view is that the concept of a digital twin is simple, but simplifications of their capability and misappropriation of the term to other disciplines threaten to curtail their potential. While they are a digital representation of the physical world, the two-way communication between the physical and digital entities is the critical component.
“It’s not just a simple representation – previously we had so many different tools and technologies to represent an object, BIM for the object and GIS representing the world around it,” he said.
“When we talk about digital twins, we are talking about something that is able to exchange data with the physical object as well.”
Dr. Sepasgozar says that a lot of actors may think or say that they are creating working with digital twins, but understanding the differences is key to their embrace and adoption.
“We can’t downgrade the digital twin to all current digital practices, we need to fully understand it first. If we know exactly what a digital twin is, companies can appropriately set their strategy towards achieving them, which will in turn affect the adoption rate,” he said.
He likens a digital twin to a ‘killer app’, a technological zenith such as connected taxi services such as Uber, Grab or DiDi – a nexus of physical and digital with a human at its centre. Although novel in its own right, Uber could not have been realised without the process of miniaturisation that brought us proliferous smartphones, AI, Google Maps, and positioning technology. Dr. Sepasgozar argues that the relationship of digital twins to other spatial modelling disciplines and developments such as cloud computing and machine learning can be viewed similarly – and like Uber, the sheer power and usability of implementation that puts the user in such an empowered position will drive their success. All actors stand to benefit, from labourers on site to residents, asset owners and users.
“We need an environment to bring the physical and digital twin together, in which they interact continuously – and the human is at the centre of that process, benefitting from this communication between physical and digital representation,” he said.
“So this is a lot more than sitting in the office and playing with data. It is real time, useful and solves problems.”
Regional momentum
This description of using a fully-fledged digital twin – of previously untapped power and insight, related with excitement – came up again in a conversation with one of the leading lights powering the use of digital twins in the region, Melissa Harris. Ms. Harris is the executive director of Strategic Land Assessment & Information at the Department of Environment, Land, Water and Planning Victoria and Deputy Chair of the Australia-New Zealand Spatial Information Council (ANZLIC). Both of these organisations are doing pioneering work in Australia, promoting cutting-edge projects and establishing frameworks to promote the use of digital twins, respectively.
Ms. Harris says that this tangible, user-focused augmented experience of interacting with a spatially-enabled digital twin forms a kind of innate value proposition to a range of audiences – an aspect that’s attracting supporters for her department’s initiative to further develop critical geospatial datasets and capabilities for the state.
“One of the great things about digital twins is that they provide a really rich experience for people interacting with them, whether they’re from a technology background or not. We’re finding that’s really powerful in terms of building the case to support and invest,” she said.
DELWP is developing a spatially-enabled digital twin for Melbourne’s Fisherman’s Bend precinct, the largest urban renewal project in Australia, covering 480 hectares over two municipalities and linking Melbourne’s CBD to the bay. The term ‘spatially enabled’ indicates that the twin confers to a set of principles developed by ANZLIC in collaboration with CSIRO’s Data61 and Smart Cities Council Australia and New Zealand (SCCANZ), ensuring that it conforms to current 3D data standards, and contains integrated temporal data (4D) and foundational spatial datasets – particularly cadastral, parcel and land use data. This project contains a range of other goodies too – 4D datasets created by machine learning algorithms from archival imagery of the precinct, BIM models of buildings, live feeds from sensors to monitor traffic and transport data.
Ms. Harris says that for a planning use case, the essence and power of a digital twin is the convergence of an array of datasets into a unified view – in 3D and 4D that would otherwise be discrete, locked up in siloed government and private databases.
“When it’s at your fingertips in a single platform, and you can use many different combinations to run queries and scenarios and model different outcomes, you have vastly superior decision making capabilities that you wouldn’t otherwise have,” she said.
“Some of the really interesting opportunities these platforms provide us have the potential to revolutionise government decision making – and how we consult with communities about those decisions – and in doing so, really make a contribution towards those long term government goals of liveable, sustainable places and communities.”
The Fisherman’s Bend digital twin evolved from a need to investigate 3D and 4D formats in a live environment – a component of the ongoing work to modernise the state’s core geospatial and planning datasets that have been developed over decades. Ms. Harris said that her team learnt so much from the development and use of the twin that it’s established a new set of strategic goals for the ongoing work on developing the foundational datasets from planning and cadastral data in legacy data stores.
“Despite so much technology and innovation over decades, digital workflows are still very much characterised by 2D, even analogue formats,” she said.
“But we have learned so many fabulous things – we’ve now got to think about how we can scale these critically important data sets over the near future, to interoperable standards-based 3D and 4D formats that support this kind of capability at scale.”
Watch this space
While digital twins in their fullest sense are in their nascency in our region, momentum is building. ANZLIC has identified digital twins and smart cities as one of its key initiatives for maximising the use of Australia and New Zealand spatial information. Initiatives to collaborate on agreed terminology, standards for interoperability and data exchange are well underway.
FrontierSI is preparing to co-invest in three areas to advance creation of digital twin products and services in support of its government partners’ priorities, and has developed a collaborative partner model that includes development and deployment phases of digital twins – accounting for governance arrangements for data acquisition, sharing, licensing and access and fusion. NSW Spatial services launched its huge digital twin project in March 2020, and the recently formed Digital Twin Consortium is developing its draft charter. And while cloud computing, machine learning, UAVs and reality capture advancements have brought about a potential new paradigm, we are only just witnessing the emergence of this technique. Watch this space.
Stay up to date by getting stories like this delivered to your mailbox.
Sign up to receive our free weekly Spatial Source newsletter.