I’ll give you one of the definitions I liked the most: “The digital twin is the virtual representation of a physical object or system across its life-cycle. It uses real-time data and other sources to enable learning, reasoning, and dynamically re-calibrating for improved decision making”. Digital twin concept is present since 2003. and it was used by NASA for pairing of technologies for its Apollo missions.
Idea behind digital twins is to make sure that outages and system breakdowns are reduced to the bare minimum. How to achieve this? Well, this digital replica uses software analytics for all the historical and live data gathered from IoT sensors of the original. Those findings, combined with the knowledge of IT/OT experts, represents the base for machine learning as a next step and AI (Artificial Intelligence) at the end.
Even though we often bind digital twins with manufacturing this will change in a very soon future as more and more things become digital and connected, since we have enormous number of different IoT sensors on the market now. This will basically give us a possibility to create digital twin models for all kind of assets (houses, buildings, etc.). Those digital twins will be used as a base for numerous what if integration projects (on for example city level scale) – so we’ll be rather precise about what elements are needed, where, when. At the end of the day we’ll have information about time scale and needed budget for getting fully functional physical deployment of some unthinkably complex solution.
For now greatest adaptation of digital twins would be a designing and pre-production digital testing of of aircraft engines, wind turbines, offshore platforms, trains, HAVC control systems, buildings, utilities (electric, gas, water). Another application would be a visualization for remote troubleshooting and pre-production fix testings where we’d get proof of solution within digital simulation. IoT sensors are playing important role here, since they are showing behavior of a physical device in a real world and also provide a real time data as well as historical data through a device life cycle.
Why are digital twins important? Well, beside further development of the product and improvement of its quality, they also bring higher efficiency, new aaS (as a Service) business models, they bring add on services and improve very basic and most important thing for any kind of business – customer satisfaction.
Not everything is so shiny in a world of digital twins, since for some projects they can make things over complicated and create additional cost. Additionally, there are projects where we have to be very careful with privacy concerns and of course there are also security concerns.
So, for a conclusion, I’d just state that Gartner predicts that by 2021. half of all industrial companies will use digital twins and see an average efficiency gain of 10% from their usage. We’ll see how precise Gartner was with its prediction.