A primer on digital twins in the IoT
The emergence of the IoT has led to an ever more heightened interest in the digital twins concept. The research firm Gartner, for example, described it in 2017 as a “disruptive trend that will have increasingly broad and deep impact over the next five years and beyond”. What is this concept all about?
IoT and digital twins go together perfectly
The digital twins concept is not new. Digital representations of physical objects have been used for decades. Take NASA for example: They were not able to see and monitor their systems physically because they were too far away, so they created digital models to simulate and analyze their systems back on earth.
In the IoT, a digital twin can provide a holistic view of most – if not all – capabilities an asset has. This implies that it helps to orchestrate different aspects of an IoT device. By providing a unified model and API, the digital twin makes working with IoT assets very easy. A digital twin is the representation of the thing itself and the contact point to access and work with different capabilities and features of that thing.
The major benefit of digital twins in the IoT is that you do not have to worry about connecting to the asset to extract and transmit any data. Instead, you can simply deploy applications in a secure sandbox in the cloud, which works with the digital twins as if they are sitting side-by-side these physically distributed IoT assets.
This sandbox approach reduces security risks because the applications are not deployed on the asset, but only in the cloud. Ultimately, development costs are reduced, which means IoT applications can be developed faster. Thus, digital twins in the cloud offer the potential to open up many new functions and solutions.
Use Cases for digital twins in the IoT
The fields of application for digital twins are manifold and not confined to a specific industry or area. They can be used for a wide range of scenarios. On one hand, there are digital twins that just model and represent a single sensor within a device. On the other, there are digital twins that reflect many aspects of a whole campus of buildings, comprising energy, usage, topology, and more.
Digital twins can, for example, be implemented in the sphere of the Industrial Internet of Things (IIoT). Take the manufacturing process as a general example. By equipping machines with sensors, you can collect a variety of operational data – not just relating to the behavior of the machine itself but also on the conditions in the factory as a whole. By using a digital twin, you can consolidate and analyze these data sets as well as replicate production processes in the virtual world. Over time, if deviations in performance become apparent, manufacturers can take action to optimize their production processes.
There are use cases outside manufacturing and Industry 4.0, such as a digital twin put to use in the context of a connected building. Here you can simulate how a building is used, based on historical or comparative data, and test changes in the building’s design, or the digital twin can call attention to rooms that are wasting energy or are used only rarely.
Usage-based insurance is another possible use case: Instead of having to deploy an expensive telematics unit on each new customer’s car, the application can now simply be deployed in the cloud. The digital twin can be used to calculate the driver’s individual driving score in real time.
Important considerations before implementing digital twins
Now that digital twins are in the limelight as a disruptive trend, one shouldn’t forget that implementing them isn’t always easy. An important pillar of using digital twins is data. Gaining access to this data across the respective organization can be an arduous and frustrating process. This hurdle can only be overcome if companies adjust their processes and systems. They have to break down silos and make data easily accessible across the organization. A holistic approach to storing, managing and manipulating data is essential, but this is often easier said than done.
Another challenge is the complexity of potential use cases. Collecting data is one thing but evaluating and using it is another issue entirely. Digital twins cannot be fully implemented overnight. It takes small steps to find out where a digital twin can create value, and what its other benefits are.
The digital twins concept is bound to evolve in the coming years, as the variety of possible applications grows, and companies continue to evaluate use cases. Where is this concept going? It is likely that the adoption of digital twins will grow as companies find out where they can be beneficial. In the years ahead, we at Bosch.IO expect greater customer demand for the virtual modeling of twins, including semantics and simulation.