What is a Digital Twin?

In this guest blog we hear about the use and benefits of a digital twin. 

 

A digital twin is a virtual representation of an IoT device, including its elements and dynamics. It represents how the IoT device functions throughout its lifecycle, understanding the elements that it is composed of, their dynamics, and the dynamics of how those elements are put together. The digital twin uses real-world data as well as simulations and data analysis. These can be used to answer hypothetical questions and can intuitively present these answers.

The virtual representation of a digital twin can take many different forms (or, preferably, an extensive combination of them), such as:

  • Bill of materials
  • Manuals
  • Job plans
  • Fault codes
  • 2D or 3D CAD files
  • AR/VR models
  • Maintenance plans

 

 

It is of utmost importance that the digital twin represents its physical counterpart as best as possible. Therefore, each physical twin will have its unique digital twin, which will have all information of this asset, including its history and context. The same asset in a different environment regarding humidity and temperature may require a substantially different maintenance plan. Another example could be an asset that can be used in different industries, with different dynamics of use. The same asset can have a very different digital twin in each industry. In one industry an emphasis may be on maintenance plans, bills of materials, and the impact on other assets. Whereas in another industry safety plays a larger role, and health and safety information is far more relevant for the digital twin.

What is the use of a digital twin?

The purpose of a digital twin is to simulate real-world systems, to help people make better decisions that impact the real world. The digital twin is used for learning, understanding, and reasoning. A digital twin can be used to answer what-if questions by running simulations, and the results should provide insightful and intuitive answers.

The digital twin influences all phases of your process: design, realisation, and operation. In the design phase, the digital twin can be used to gain operational insight which can lead to an improved design. In this phase, physical elements, such as a Bill of Materials, or virtual elements, such as chips can shape the digital twin.

In the realisation phase, the digital twin can be used for continuous learning, to result in better manufacturing.

In the operational phase, the digital twin can assist in informed services and support, and improve manufacturing. The corrosion and wear from the use of your asset should be fed back to the digital twin. Sensors from the physical twin can provide this feedback to meters in Maximo, such as meters for run hours, temperature, vibrations, etc. This information in your digital twin can then be used for analysis and predictions for future use.

Consider the following scenario; a pump may be showing signs of a probable malfunction in the near future, which requires spare parts for reparation. The spare parts may take some time to be shipped to the asset location, resulting in the pump operating suboptimally for that period. The digital twin can be utilised to calculate the impact of the scenario on the performance of the pump, that way an effective call to action can be put in place highlighting which potential options available to receive the new part will be the most cost-effective for the business.

System of systems

In the example above, the focus of the digital twin is on the system of one asset – “the pump”. However, if all assets and parts of a larger system are created as a digital twin, we speak of a system of systems. If the digital twin is set up as a system of systems, meaning the pump itself is set up as a digital twin, but so is the larger asset the pump is a part of, the impact on the larger system can be predicted, and even more extensive analysis can be created.

Steps

It takes three steps to make use of a digital twin:

  1. Data needs to enter the system: Using sensors, real time data is brought in.
  2. Cognitive computing: Data analysis is performed to make sense of the data.
  3. Recalibrate the environment: Dynamically recalibrate to adjust changes and processes.

Overall, the goal of a digital twin is to have real-time, predictive analytics in every step, from design, realisation, and operations. The more detailed the data in your virtual twin, and the better the integration with its physical twin, the more valuable and reliable predictions and analyses can be overall.

Find out more about IBM Digital Twin Exchange here