Nuances of a Digital Twin

Henry Wheeler Shaw had said, ‘There are two things in life for which we are never truly prepared: Twins!’.  In contrast to the humor writer’s opinion, the Digital Twin has been welcomed with more than open arms by the industry. 

A 2019 survey by Gartner, mentioned 13% of the organizations executing IoT projects had already applied the digital twin technology in their operations. The same study also predicted that by end of the year 2020, that figure would jump to a whopping 75%. In a world where adoption of latest technology for operational effectiveness has become necessary, these figures alone should be enough to sway the minds of even the orthodox. In a cut throat industry where one needs to either adapt or get left behind it becomes imperative to get familiarized to and implement relevant technologies. 

To do that, one must understand its nuances.

In our last blog we covered how digitalization gave rise to the digital twin concept and a basic level of understanding about the technology. Elaborating on it further a comparison to similar tools and technologies is the best way to better understand the subtleties of the digital twin. 

For example, one would argue that a CAD model is in a way a ‘digital’ replica of the asset in question and a simulation run with it is a merger of the physical and the virtual analysis. While the ‘digital’ replica is common to both, the similarity stops there. Real-time data provided by the sensors on the system is what is essential for the digital twin which no amount of CAD modeling can provide. 

Coming to the digital models used in simulation, the data inputs required for it are generated (at times manipulated) and are not available in real-time. In short, a simulation is no digital twin though a digital twin can be used for a simulation. 

The bigger confusion in today’s world is when people treat telematics as a digital twin. Telematics is a term used for devices or systems which transmit data using wireless technologies. So in theory, the real-time data one gets from telematics makes a virtual replica of the system. What sets it apart is how the data is populated for the digital twin. This is where the true difference comes to fore.

Data for a digital twin is updated from not one but multiple sources. The data from the individual component along with the way it behaves with other components in a state of harmony. This allows it to detect anomalies even before it occurs making the model predictive. This is what sets them apart from other virtual models. While it can be used in a wide array of industries, let us consider the case of telematics for vehicle fleets and its management to better understand the difference.  

What telematics enables, is to remotely acquire the entire data of the vehicle from component faults to driving behavior, live tracking etc. While all this is important, wouldn’t predictive data be better?! Getting an alert for a vehicle fault code is important but it leaves the fleet manager helpless when the vehicle is on the move. Wouldn’t the same information be more useful as a prognostic alert so a pre-emptive check can be done and any possible downtime of the vehicle avoided all together? To make the deal even sweeter, what if the alert came with the repair strategy? So instead of starting from scratch to solve the problem, the person just needs to check the possible causes highlighted by the system and voilà.

This is exactly why, we at Intangles believe that the concept of a connected vehicle is no longer a USP but has become a part of the core product. A good telematics system is essential to make the digital twin of the vehicle smarter but relying only on telematics to manage a fleet is no longer prudent. Digital twin solution is the augmented service offering which only a few are able to provide.