A Digital Twin is an exact digital (or virtual) replica of something that exists in the physical world. Using a bridge between the physical and digital, a “digital thread” of data is transmitted seamlessly allowing the virtual entity to exist and operate simultaneously with the physical. A Digital Twin is the future of smart asset management in the operation and maintenance phase.
Why do this? There are many benefits - running safe simulations, asset performance optimisation, risk management, remote inspections, customer experience and cost reduction are just a few headlines.
Often Digital Twins are confused with BIM/BMS systems. Whilst there is some overlap, I see the distinction in two ways: firstly, a Digital Twin uses a model (could be 2D or 3D) as part of the user interface to visualise the data, whereas BMS systems are far more basic and generally use simply process maps; secondly the the spectrum of data inputs is much greater in a Digital Twin and there will be more useful outputs or connections into other software.
Most of the time, at the end of construction many people would be forgiven for thinking that the Asset Management team receive an all singing, all dancing magical BIM model, however more often than not the reality is that this is just a relatively ‘dumb’ 3D model with little capability for asset or facility management.
George Mokhtar at T&T describes creation of a Digital Twin for a typical building project as a relay race:
Creating a Digital Twin is just the first part of a relay race where the Digital Twin baton must be passed from the Construction Team to the Asset Management team. All too often, the baton is simply dropped at PC or the receiving team are not given the training or budget to be able to keep hold of the baton and use it going forwards.
So the key to creating a successful digital twin is knowing what benefits it will bring, how/when it will be used, and ensuring these things are constantly in the mind of those creating it. The value often comes later, but you have to build it right. The other important factor is to recognise that running a Digital Twin incurs a cost during use, so you need your incoming Asset Management team on-board long before you hand them that baton.
If all works to plan the Asset Managers who could have received that ‘dumb’ 3D model, now receive a well thought through, Digital Twin that is synchronised with the physical asset and provides meaningful monitoring and optimisation. And the best bit… they were expecting it, have been trained and have budget to maintain the model and the software that sits behind it.
Whilst I certainly think of Digital Twins as being associated with single buildings post-PC, there are some other use cases to consider. A Digital Twin has the capability to span the full lifecycle with users in all phases able to benefit from it - for example a Digital Twin of the operations on a construction site could help with energy usage, logistics or progress. And if you have a number of buildings on an estate or portfolio, there could be a higher-level Digital Twin which receives information from each building - just like how a 3D model is used to visualise the data, in this instance the data might be interrogated on a map supported by an interactive dashboard display.
If you read my recent newsletters on Intelligent Buildings (1 and 2), you’ll recall that a Digital Twin is a big part of what makes a building intelligent - this is where a lot of the data comes together. A Digital Twin is one of those key ‘enablers’ that is important to get right at the start. From there, further functionality can be incorporated.
Currently, Digital Twins that exist are still fairly crude with 90+% of them being backward looking, pulling largely static data and only providing limited operational assistance (e.g. Planned and Preventative Maintenance). Only a small few are pushing into the territory of improving operational efficiency. That leaves a big gap to be filled in the future where they actively respond, control and then predict and run ‘what if’ scenarios automatically.
LinkedIn linkedin.com/company/stevesnewsletter
Twitter twitter.com/SteveNewsletter