top of page
  • Writer's pictureSafer Highways

Digital Twin – Hip or Hype?



Digital twins are flooding tech trend searches, presented as if heaven was raining them down like manna—they are the solution to everything from productivity to sustainability. But are they really all they’re hyped up to be? And, if so, how does one get one? The first step is looking at the art of the possible.


Using digital twins for planning makes it easier to see the future and engage with stakeholders to address constraints, build scenarios and rapidly ingest opinions to test new ideas. Where previously constrained by a drawing or a painstakingly built replica, now interaction with a digital twin of the proposed building, road or entire city eases the decision-making process. The use of digital twins during the planning stage can better inform the design and increase the chances of achieving the intended outcomes.

Using digital twins for design helps optimize designs by testing everything from buildability to how well different materials stand up to various climate conditions. Where schedule or budget previously limited the number of design iterations possible, now hundreds of iterations can be run based on parameter inputs, leveraging technology like generative design to automate options far beyond our imaginations.

Using digital twins for construction improves productivity and shortens delivery in digital rehearsals. Linking schedule and cost to design elements and rehearsing the construction sequence optimizes logistics, equipment and even safety. Simulating facility operation during construction can lead to enormous savings in operating costs by reducing downtime and disruption.

Digital twins for operations and maintenance, according to some, is where the digital twin was originally conceived—the true “building model” of the 1980s brought to desktops by ArchiCAD in 1987. The 3D representations may have been the first digital twins, but not like today’s sensors from connected data ecosystems which monitor, measure and react, sometimes autonomously, to conditions affecting optimal performance.

What all digital twins have in common is data, generally lots of data, and a connected data ecosystem where computer logic can be applied to predict outcomes. The algorithms can be continuously improved through training to improve the quality of the predictions—this is known as machine learning, a key aspect to unlocking the digital twin superpowers.

Technologists certainly know what is possible. But the real secret lies in the owners and operators that know what’s needed. Quite the chasm often exists between these two groups, either because the operators can’t translate what they need for the technologists to understand or sometimes because the technologists don’t appreciate what they don’t know.


While owners and operators are already challenged by the chasm in realizing digital twin potential, now they have just been dealt another blow—the COVID-19 pandemic. Owners and developers are looking to keep projects on track, but on track for what? Will user demands change? Do input factors need to be adjusted? If ever there was a time for a digital twin, it’s now. The ability to run simulations quickly based on new projections, volumes or funding has the potential to support better decision making to get projects back on track or to pivot to address a new track all together.

Deciding where to start and what data is needed is determined with the end in mind—the end result for which the organization or the project is striving, not one individual’s idea of the end. This is often where the story goes wrong. Owners and operators challenged by the data chasm are expecting someone to roll out a recipe with all the ingredients listed. However, the ingredients are predicated by the end result the particular owner has in mind, which can differ dramatically from owner to owner and from project to project. Defining the end result and applying operational, construction, design and engineering or planning expertise, determines the data requirements.

The design life cycle is typically examined in order of its events, but for a digital twin, the thinking is reversed, starting with the end and working backward.

In the end, there are a lot of ways to get from A to B—starting from scratch and building the twin over time, or piecing together data already existing and bringing it together into a common environment, or even, without the luxury of a clean slate or the time to collect and merge disparate data, today’s reality capture technology is accessible and affordable. LiDAR or 3D laser scanning technology, for example, are both effective means of reality capture. The scans can be quickly translated into as-built 3D models which, when combined with 2D as-built drawing data, can have a digital twin ready in no time. However, keeping the end in mind still remains the constant reference. Laser scans can produce enormous amounts of data and one must decide what to translate from the scan into the model or get buried under the weight of the new twin. Once the initial investment in the scanning is complete, the necessary data can be extracted when needed, adding layers of details to support discrete functions and perform specific analysis over time.


This article was originally published on constructionexecutive.com, October 3, 2020.

30 views0 comments

Hozzászólások


Recent Blog Posts

NEWS AND UPDATES

bottom of page