Digital technologies are no longer considered tools to satisfy a need for efficiency, they are active agents in value creation and new value propositions [1].
The term “digital twin” has entered the regular vocabulary across a myriad of sectors. It’s consistently used as an example of industry revolution and is considered fundamental to transformation, but the broad scope of the concept makes a common definition difficult. Yet it’s only once we understand and demystify the idea – and can see a path to making it reality – that we will start to realise the benefits.
Heavy promotion by technology and service providers has inflated expectations, with most focusing on what a digital twin can potentially achieve when fully implemented, which is like buying a unicorn even if currently cost-prohibitive. Few refer to the milestones along the journey, or incremental value-proving developments. This is evidenced, in part, by the fact that only 5% of enterprises have started implementing digital twins, and less than 1% of assets have one [2].
Over the course of three blogs, I will attempt to demystify the concept and break through the platitudes, answering the fundamental questions: What is a digital twin? What type of new skills and capabilities are required? Will a digital twin generate value? And will it support better decision making?
“Digital” in context
Digital twins are symptomatic of the broader trend toward digitalisation, which is having a profound effect on businesses and society. Widely cited as the “fourth industrial revolution” [3] or Industry 4.0 (broadly following: steam power (c1760-c1840), electricity (c1870-c1914) and microchips (c1970)), it’s characterized by a fusion of technologies that blur the lines between the physical, digital, and biological spheres – such as artificial intelligence, robotics, autonomous vehicles and Internet of Things (IoT).
Though the exact dates of the earlier revolutions are disputed, their timeframes were slower than the rapid pace and scale of today’s disruption, and still they saw companies and individuals that were slow or reluctant to embrace change being left behind.
The digital revolution is unique, and derives in part from a new ability to massively improve quality and productivity by converging technologies and sources of data within a collaborative framework, which inherently challenges the business and organisational models of the past. Not only this, but the online connection of all assets together (the Internet of Things), is the key enabler to the next phase of industrial development.
The complexity of assets, and cost of developing and operating them makes any promise of efficiency gains and improved performance immensely attractive. However, the reality of digital transformation to offer these rewards has too often fallen short. The failure comes from a rush to introduce digital technologies, products, and services without understanding the work processes in which they will be used, or the associated behaviours and joined up thinking required to make them effective.
While individual products and services have their place, significant gains in efficiency and productivity will only come by weaving a constellation of technologies together and connecting them with data sources, followed by supporting management and application of that data through project, asset and organisational developments.
Is data the “new oil” or the “new asbestos”? and how can industry start tangibly benefiting from the digital twin concept?
With data apparently the “new oil”, or maybe the “new asbestos”, and against a backdrop of digital transformation being viewed by many sceptics as a fashionable buzzword, how can industry start tangibly executing and harnessing the benefits of the digital twin concept?
Digital twin basics
Fundamentally, a digital twin is just a digital representation (model) of a physical thing – its ‘twin’; and therein lies the complexity of this industry agnostic concept. Other commonly used terms, such as Building Information Modelling (BIM), Building Lifecycle Management (BLM) and Product Lifecycle Management (PLM) represent similar concepts with some important distinctions, that are all part of the same theme of data generation and information management.
The term “digital twin” first appeared in 2010, developing from the conceptual evolution of PLM in 2002 [4]. Since then, it’s meaning has evolved from simply defining a PLM tool into an integral digital business decision assistant and an agent for new value and service creation [5]. Over time many have attempted to define the digital twin, but often these definitions focus on just a small part of the asset lifecycle, such as operations.
“A digital twin can range from a simple 2D or 3D model with a basic level of detail, to a fully integrated model of an entire facility with each component dynamically linked to engineering, construction, and operational data”
A digital twin can range from a simple 2D or 3D model of a local component, with a basic level of detail, all the way to a fully integrated and highly accurate model of an asset, an entire facility, or even a country [6], with each component dynamically linked to engineering, construction, and operational data.
There is no single solution or platform used to provide a digital twin, just as there isn’t one CAD package used to create a drawing or 3D model. It’s a process and methodology, not a technology; a concept of leveraging experience-based wisdom by managing and manipulating a multitude of datasets.
While a fully developed digital model of a facility remains an objective, practically speaking, we are delivering only the “low hanging fruit” pieces of this concept for most facilities now. These fractional elements, however, all point towards a common goal: to contribute a value-added piece that is consistent with the overall concept of the digital twin. As technology and techniques improve, we predict the convergence of the individual parts and the emergence of much more complete digital twins for industrial scale facilities, and ultimately entire countries.
“There is no single solution or platform used to provide a digital twin, just as there isn’t one CAD package used to create a drawing or 3D model”
The ultimate aim is to create a “single version of truth” for an asset, where all data can be accessed and viewed throughout the design-build-operate lifecycle. This is distinctly different to a “single source of truth”, as a digital twin is about using a constellation, or ecosystem, of technologies that work and connect.
The digital twin promises more effective asset design, project execution, and facility operations by dynamically integrating data and information throughout the asset lifecycle to achieve short and long-term efficiency and productivity gains.
As such, there is an intrinsic link between the digital twin and all the ‘technologies’ of the fourth industrial revolution, principally IoT, artificial intelligence and machine learning. As sensors further connect our physical world together, monitoring the state and condition, the digital twin can be considered the point of convergence of the internet-era technologies, and has been made possible by their maturity. For example, the reducing costs of storage, sensors and data capture, and the abundance of processing power and connectivity.
The digital twin is a data resource that can improve design of a new facility or to understand the condition of an existing asset, to verify the as-built situation, run ‘what if’ simulations and scenarios, or provide a digital snapshot for future works. This vastly reduces the potential for errors and discontinuity present in more traditional methods of information management.
As asset owners pivot away from document silos and toward dynamic and integrated data systems, the digital twin should be become an embedded part of the enterprise. Like the financial or HR systems that we expect to be dynamic and accurate, the digital twin should represent a living as-built representation of the operating asset, standing ready at all times to deliver value to the business.
Each digital twin fits into the organisation’s overall digital ecosystem like a jigsaw, alongside potentially many other digital twins for different assets or systems. These can be ‘federated’ or connected via securely shared data – making interoperability and data governance key. In simple terms, this overall digital ecosystem consists of all the organisational and operational systems, providing a so-called ‘digital thread’.
Author: Simon Evans. Digital Energy Leader, Arup. Delivery Team Lead, National Digital Twin Programme
[1] Herterich, M. M., Eck, A., and Uebernickel, F. (2016). Exploring how digitized products enable industrial service innovation. 24th European Conference on Information Systems; 1–17.[2] Gartner, Hype Cycle for O&G [3] https://www.weforum.org/agenda/2016/01/digital-disruption-has-only-just-begun/ [4] Digital Twin: Manufacturing Excellence through Virtual Factory Replication. White Paper, pages 1 – 7 [5] Service business model innovation: the digital twin technology [6] Centre of Digital Build Britain, The Gemini Principles
Leave a comment