What is a Digital Twin?
That seems to be the question of the year. What exactly is a digital twin? I’m asked that question almost as much as I ask the same question myself. After about a year of trying to dive into the subject I didn’t come up with a lot of answers, but drew a lot of honest questions.
A BIM Rebranding?
Yep. It’s a rebranding of BIM. Autodesk simply can’t seem to, or won’t produce a viable interface between Revit, Navisworks, Civil 3D (and whatever else they acquired that month) to a typical end user who could actually use the data for facility management. I think the term BIM is finally running out of marketability. They’ve already made word salad out of what the B the I and the M actually means. I’m a BIM Director and I can’t explain to people what my title means. Why not just rebrand something instead of coming up with something solid and new? Piecing together hundreds of Revit models for huge buildings with multiple projects going on over the span of years is, in almost every instance, undoable. Ever upgrade a Revit model and have a warning saying that 300 taps must be attached to a duct? I wonder if that’s going to cause any issues. I don’t want to be the person having to deal with that. Neither does anyone else. Another issue is, it seems everyone wants to talk about the twins but very few folks actually want to dive in and take responsibility for it. It happens to the best of us. I think the fake Pie Chart below sums up most presentations you’ll see on the subject.
So, what is a Digital Twin technically?
Over the past few years I have worked with PhD students, clients, and colleagues on that definition. It does start with BIM, but the actual digital twin part refers to the interface that people can use without being a Revit expert I was talking about. It’s about getting a 500 LOD model into an application that works for facility management. There’s lots of 500 level models out there and there’s lots of awesome facility management applications out there but the two never successfully seem to get married up. Also, a digital twin has controls appended to the assets. For example, if you click on a roof top HVAC unit, not only do you get the typical LOD 500 data but you are supposed to have live sensors monitoring the equipment, all connected by IP address and located by GPS. Pretty exciting stuff.
The logistical issue now is existing buildings. My jam is aviation and healthcare. These buildings are already there. They are added onto over the years and most of the critical infrastructure we would need access to, is either buried in a wall or ceiling… or actually buried underground. The effort that it takes to get an actual “twin” of what’s there would cost more to the owner than the cost savings would ever pay back.
Then there’s maintenance. Even if everything above was accomplished (they tore down the ceilings and walls and dug up the underground infrastructure) In an effort to use a digital twin to maintain a building, these maintenance models need to be constantly maintained themselves. This takes additional staff, constant training, and the willingness of the people actually doing the work to play along. Half of the staff is probably close to retiring. Trust me, there’s reluctance. I know, that’s exactly why we need to capture this data now, but you might want to work on your patience if that’s an issue for you.
How do we get there?
If you know me or have read my stuff before you probably know I act tough but usually back off. I spent my career on this subject in the trenches so I see the ugly more than most. But I do believe we can get there. It does start with making sure your design firms are giving you what you need upon project close out. That means the up-front bid documents need to have a solid BIM Execution plan and the deliverables need to match your expectations. Constructing a new building out in the middle of a field somewhere helps also, but that’s not generally the case.
One thing I also like to point out, is you don’t need to go from paper drawings to a digital twin overnight. Looking at the end game while assessing where you are at now is cause to say “there’s no way on earth is this going to happen”. Start from square one. Establish what you do have in place and build from that. Get management backing and for crying out loud learn as much about the subject and the software as you possible can. Be a leader.
Please. Get a hold of me if you need another pie chart, completely agree, or if you think completely differently about the subject. You can find me on LinkedIn.