Digital twins help project teams and facility operators build and operate better, more sustainable assets through the aggregation of facility data, the development of visual intelligence (VQ) models, and the generation of real-time insights.
A digital twin is a virtual representation of a physical object or system. It is created using 3D models, sensors, manufacturing data, operating data, other data sources, and specialized software to effectively capture and visualize the facility or asset's characteristics and behavior. The digital twin can be used for a variety of purposes, including simulating how the physical object or system will behave in different scenarios, optimizing its performance, and predicting and preventing potential issues.
A digital twin is a process that improves a product, rather than a product itself. A 3D model is not a digital twin. Building Information Modeling (BIM) is a design methodology which includes the development of 3D geometry and associated metadata; BIM, or specifically 3D modelling is often conflated or confused with digital twinning; they are not the same.
3D Models, Digital Shadows, and Digital Twins
The key differentiator between a digital twin and a 3D model is the feedback loop that exists between the physical asset and the digital one, which is factored in real-time. This feedback loop, through the use of cloud technology, must bring data from the physical asset into the digital twin; in turn, the digital twin must analyze the data and action it, interfacing directly with the physical asset to affect a change in system operations or configuration.
When data is not factored in real-time, or there is a manual interface within the feedback loop, the digital asset is referred to as a digital shadow. A digital shadow is similar to a digital twin, but is not fully automated and requires manual (human) intervention. This intervention may include manual data input or manual physical asset interface. The benefits of a digital shadow may not be as great as a digital twin, but a digital shadow also requires less hardware and less expense to both build and operate.
Everyday Digital Twins
A common example of a highly accessible digital twin is a smart home system. A smart thermostat can provide feedback on energy use; it can tailor the environment of your home based on a variety of predefined metrics. You can control your asset (furnace, lights, or security devices) from any internet-enabled hotspot in the world, changing the metrics to suit your needs. The data collected by the system enables the homeowner to make informed decisions that can impact their personal comfort and energy use, while also monitoring the entire system for potential failures and upcoming maintenance. While this digital twin has no 3D geometry, it enables a physical-virtual feedback loop to creating a better living environment based on live and historical data. The smart thermostat is, therefore, a concrete example of an everyday digital twin.
Asset Digital Twins
A more complex digital twin example is a digital twin for a commercial facility. A BIM model has been developed for this facility, as well as a plethora of other data sets. These data points may include manufacturer model numbers for sub-assets (pumps, doors, furnaces), installation records, maintenance data, and service manuals. Within this facility digital twin, this information is combined using specialized software. The facility is virtualized; every physical component has a virtual existence; this may include 3D virtual geometry and corresponding metadata, or non-3D asset data points which are intricately architected to support data visualization, automation, and machine learning / artificial intelligence.
Next, the digital twin integrates real-time facility data. This data is recorded through the use of sensors within live operating systems; these sensors provide users with real-time operating data. The digital twin, through analytics and artificial intelligence, automatically makes system modifications based on operating data and desired system characteristics. Much like our smart thermostat, when readings are out of tolerance (temperatures are too high or a component has failed), the digital twin will take immediate action to address the fault and return the system to tolerance.
While the complexity of this digital twin is much greater, the functionality remains similar to the smart home example. Digital twins may be deployed for an asset or a full facility; the more intricate the physical objects, the more complex the digital twin will be. Through the visual intelligence (VQ) the digital twin provides, operators can navigate with a single, trusted source of asset information.
As mentioned above digital shadows do not require full system automation. Therefore, teams may gradually evolve a digital twin through the generation of facility digital shadows. While less complex, digital shadows still generate immense value for project and operations teams alike; consider digital shadows to be stepping stones to digital twins.
Process, Not Product
By approaching a digital twin as a process rather than a product, teams focus on quantifiable goals and efficiencies that benefit all stakeholders, rather than the delivery of a functioning model. Reframing the digital twin as a process helps teams effectively identify the business goals for digital twinning, determine how a digital twin will help achieve these goals, and craft a process that delivers on the objectives. When process is the driver, teams will make informed investment decisions and are far more likely to achieve value from their digital twin than those teams who approach digital twinning as a product deliverable activity.
A digital twin is iterative; it enables collaboration, optimization, and innovation. Handing over a digital twin with no future plans to maintain or improve the twin defeats the purpose of the process. A digital twin is not a set of drawings, a render, or a point cloud; it is a solution that empowers teams to interact with, and improve on, an asset’s operations throughout its lifecycle. Defining a digital twin as a product is therefore detrimental to the teams that must work with it, as it implies that the digital twin is at some point complete and fit for use. A digital twin is fluid, adaptive, and routinely updated.
What Can a Digital Twin Do?
If you need to determine the functionalities that define the success of a digital twin for your organization, it’s important to know what a digital twin can do. At its core, a digital twin is a singular source of project/asset truth. By providing a consistent source of trusted information, teams eliminate Dark Data, or data that is produced but which does not create value. With less time spent looking for accurate information, teams can begin working with data and visualizations to perform value-added tasks. While a digital twin can expand almost infinitely to cover any use case a team might want to explore, here is a common set of use cases that are driven, enabled, or enhanced by digital twins:
Digital Build Workflows: With the complexity of construction projects constantly increasing, teams are looking for ways to ensure projects come together on time and on budget. Teams utilize BIM models to make key decisions during design to deliver a better facility configuration. Beyond design, procurement and construction teams build on and utilize these design models onsite, producing a plethora of data sets which are integrated with the design model using specialized digital twin software. Throughout the build process, the digital twin is in development; but, project teams can leverage the digital shadow to visualize data and make informed project decisions.
Simulations: The digital twin should be an exact duplicate of reality. When working in a digital realm, we can perform simulations to understand how changes to the digital asset could impact the real-life asset. These simulations can range in complexity from system efficiency analysis, flow and output simulations, to geo-referenced sun and shadow pathing. Simulations in a digital twin are non-destructive, iterative, and informative. These simulations allow teams to understand how changes can impact their asset virtually before changes or adjustments are made to the physical asset. Pixels are cheaper than bricks; teams should use their digital twin to explore as many simulations as possible before sanctioning asset work.
Operations and Maintenance: Operations and maintenance teams can extract and action intelligence from a digital twin during their daily routines. Sensor feedback data can help alert operational staff to potential issues, flag underperforming assets, and help develop predictive processes with historical and real-time data. When facility owners look to implement a digital twin, asset operation improvement is where they will likely find the most immediate return on investment.
Sensor-Based Highlighting: A digital twin which utilizes sensor feedback for systems can be designed in a way to automatically visualize the assets and systems that are underperforming. Teams can utilize this visualization to understand the extent of the risk, and develop maintenance procedures and damage mitigation plans.
Part Replacement: With a digital twin, teams can order replacement parts, equipment, and piping spools without having to field measure existing components. Assuring fit and expediting material ordering is a valuable process that can alleviate human error and reduce maintenance costs.
Predictive Maintenance: Having a digital duplication of a physical asset allows you to track its performance over time and improve maintenance protocols. Utilizing historical data, operations teams can begin unlocking predictive maintenance procedures that aim to enhance overall asset longevity while pinpointing assets that have reduced efficiency.
Virtual and Augmented Reality: Digital twins can be enhanced with virtual and augmented reality user experiences. Involving operations and maintenance team members in virtual reality-enabled reviews allows them to visualize their daily routines and provide feedback that can improve the design or configuration of the asset. Virtual reality helps stakeholders understand the facility on a true-to-life scale and removes the requirement for technical drawing knowledge to fully understand the configuration. Augmented reality offers a different experience and displays additional information atop of reality rather than immersing the user in the digital world. Operations teams can leverage augmented reality to visualize asset performance data while walking through the physical facility, thereby seamlessly enabling real-time insight generation.
Retrofit and Expansion: Digital twins can be created regardless of the project or facility type. While it is much easier to develop a digital twin during the design and construction of a facility, you can also create a digital twin of an existing / operating asset. The deployment of reality capture technology makes this process possible, yielding a hyper-accurate representation of the asset for further refinement.
One of the primary benefits of a digital twin is the ability to generate real-time, accurate insights into the performance and operation of a physical object or system. These insights can be especially useful in industries such as manufacturing, transportation, and energy, where even minor issues can have significant performance consequences. When teams can visualize issues, or even predict them before they transpire, they can more effectively maintain and operate the physical asset.
What Can't a Digital Twin Do?
A digital twin won't immediately revolutionize your design, construction process, or facility operations. Implementing a digital twin is a concentrated effort; it requires investment, process definition, technology architecture and deployment, and iterative development. It will not immediately create value, fix project execution issues, or completely automate facility operations.
A digital twin will not define your program goals. Teams must define the business requirements and goals for digital twinning, and then ensure that the processes support attainment of those defined outcomes. Without clear goal definition, teams may make heavy investments without realizing the values sought out.
A digital twin will not automatically save you money. The process of developing a digital twin is a substantial undertaking. Time must be invested early to ensure the processes, workflows, and data sets are defined and functional. The cost savings of a digital twin are realized over the facility or asset lifespan; these cost reductions come in many forms. A digital twin leveraged during construction or maintenance may save time onsite or reduce clashes, while a digital twin utilized in an operating refinery may increase facility output or extend maintenance intervals.
Finally, a digital twin will not guarantee team buy-in to the process. If teams are not prepared for the commitment to develop the digital twin, they may withdraw from the collaboration efforts at some point during production. This risk can be alleviated by strong road maps, training, workflows, and a team commitment to the process in support of goal achievement. Training also helps teams understand how new technologies and processes can make an employee’s daily life better, thereby increasing intrinsic rewards and personal adoption.
84% of all digital transformations fail. Digital transformation failures can create disconnects between owners, contractors, advisors, and operations teams. Digital efforts fail so frequently because teams approach digital transformation processes with an all-or-nothing attitude. Organizations can combat this risk of failure by taking small steps and proving the utility of a digital twin to increase stakeholder support, amplify employee adoption, and drive incremental value throughout a facility or asset’s lifecycle.
Human beings are generally hesitant to change, and digital twinning demands an immense amount of change at every level of the organization. If teams expect to establish a trusted digital twin, participants must be prepared for roadblocks and unforeseen challenges. A digital twin is an enabler for collaboration and provides a platform to make informed decisions; but, a team that refuses to collaborate or contribute, or to make use of the twin, can quickly derail the entire process.
Where Do I Start?
To begin, determine your digital twin business requirements and goals; then develop a plan or roadmap to achieve them. Work with your stakeholders to develop a clear digital twin implementation plan. Consult experts to help develop or validate your plan. The first step in any journey, no matter how complex, is to begin with a roadmap.
Then, focus on team alignment. Generating team alignment will bolster your digital transformation effort and your chances for success. Focusing on the needs of your team, and addressing any concerns, will increase buy-in and support throughout your journey.
Pilot, pilot, pilot. Don't plan to deploy a digital twin for an entire operating facility as your first step. Focus on a single asset or a small facility area to prove the process. Start small and scale. Once you have a viable base product, you can look to expand the capabilities of your digital twin over time.
Finally, start with a growth mindset. Digital twinning can be a complex process. Start the journey with a willingness to explore, learn, and grow. Teams that approach digital twinning with uncertainty create far more business value from the process than teams that immediately seek to find a 'one size fits most' approach. You can't order a digital twin online; you must craft it, cultivate it, and curate it. Only then will you achieve measurable value through your digital twinning program.