What is all the fuss about digital twin?
Learn the basics of digital twin and how to start optimizing your processes.
If you have not been living under a rock during the last 5 years, you’ve probably heard about the latest industrial revolution: the fourth industrial revolution or Industry 4.0. It mainly focuses on connectivity, automation, machine learning, and real-time data. Through these concepts, it is how the digital twin becomes alive.
You might already be working on it without knowing it if you have looked for new ways of optimizing your processes, considered creating 3D models of your assets or consolidated your data on the cloud.
But we are not here to talk about what you’ve already done. We are here to understand what a digital twin is and how it can help your business to generate valuable insights as well as more productive and cost efficient processes.
Let’s break down what this means.
A 3D model is not a Digital Twin
The concept started to be used early by NASA when they used an updated a combination of physical and mathematical models to rescue Apollo 13.
According to Gartner, the definition of digital twin is:
“A digital twin is a digital representation of a real-world entity or system. The implementation of a digital twin is an encapsulated software object or model that mirrors a unique physical object, process, organization, person or other abstraction. ”
But what distinguishes a digital twin from any other digital model is its connection to the physical twin.
In other words, a digital twin is a virtual representation of an asset and/or a process that you can see insights in an intuitive way and interact with it having an effect on the real asset.
How does it work?
On one side, we have our physical asset. Like a turbine or pipe or substation. And on the other side, we need to create this asset virtually in 3D.
The physical object has to be equipped with sensors. These sensors generate data about the physical object such as its position, temperature, weather conditions, or energy output.
Digital twins are designed as a two-way flow of information. So there is data coming from the sensors of the physical object updating the virtual 3D representation of the asset, and any update from the user through the system on the virtual 3D model can affect the physical asset.
It has to be real-time content for the system to be able to produce valuable reports and generate insights.
As part of it, underlying modeling and analytics are essential in order to support the decision-making process of the users. These models will help to monitor production systems, run simulations, study its performance and identify issues before they occur.
In addition, all this information needs to be consolidated on the cloud to be accessible from wherever the user is, whether is a control room or in the field.
However, all the data in the world is useless if the user doesn’t have the one piece of information they need.
How to identify the best way for our users to visualize the information?
The right solution is likely to include a mix of 2D and 3D, including:
- Real-time Dashboards on a computer
- 3D Model Viewer or Modeler on a computer or mobile device
- Reports (predictions, status results,..)
- Augmented Reality(AR) or Virtual Reality(VR) to allow the users to expand the way they interact.
There are different ways of showing the data depending on the time of need. A desktop dashboard might be useful when the user is static in a control room. However, it is easier for the employee to have a mobile device or an AR glass when he is out on the field.
Building your digital twin becomes useful to train users in immersive virtual environments doing simulations, in which technicians can get “hands-on” with the digital twin with no dangerous contact or costly components.
Also thanks to the real-time connection between the physical and virtual asset, you could work on it remotely. Imagine that you want to do an operation on a remote plant, but the plant is 1 hour away and the operation to fix the issue is time-sensitive. You could manage or automate a robot from the control room to do the operation and oversee the asset position and status.
(If you want to understand better AR and VR capabilities, check out my post What does XR mean for Enterprise?)
Where can you apply it?
Digital Twin enables faster, safer, and more precise operational decisions for assets and processes. The challenge is to identify which part of the process.
Depending on the life cycle phase chosen to implement it, we can differentiate 3 different types of digital twins:
- Design Twins: applied on the engineering & procurement phase. This is typically focused on model-based verifications, validation of designs and simulation & tests.
- Process Twins: applied on installation & commissioning. Typically focused on process optimization, like deployment & configuration.
- Operational Twins: applied on operations & maintenance. You can use this digital twin from product performance optimization, control & product re-configuration to using this digital twin for trainings & simulation.
Physically large projects (such as building a spaceship, a bridge or submarine), mechanically complex processes or manufacturing projects will for sure benefit from this technology.
This type of projects are enclosed across a wide range of industries, from Automobile, Energy & Utilities, Constructions to Engineering.
On the other hand, it has also an extreme potential when it comes to Smart Cities. It can ease all management relate dto traffic, power generation, temperatures, humidity, air conditioning, and heating in buildings, pollution…
For example, there are already cities like Shanghai or Singapur that have created a complete digital twin of their own city. The digital twin specialist 51World were able to modeled 20 landmark structures and extract data from satellites, drones and sensors to connect all in the digital version.
In conclusion…
Your new digital twin can leverage the latest technologies and all your data to take your business to the next industrial revolution.
While there is a great number of applications and industries where digital twins can bring benefits, there is no guarantee that it will be useful everywhere. Not every object or service is complex enough to need sensor data as the one digital twins require. It can also be more expensive to build the digital twin than the benefits from it.
The main challenge is to identify the pain points or bottlenecks of your processes to find smart solutions.
To recap, try focusing on:
- User-centric solution: prioritize solutions to deliver value quickly to the user.
- The right data: you don’t need the 100% of it available. Focus on leveraging data science to find the KPIs that you need.
- Scalable: start small think big.
IBM has been doing a lot of work with digital twin technologies and if you are curious to know some of our success stories, check out the IBM Digital Twin Exchange.
Also, don’t forget to let me know what you think and reach out to make the future real!
