Enterprise manufacturing intelligence through digital twin

Digital twin classification

We’ve seen that digital twins can provide a wide range of benefits for modern manufacturing organisations, but different types of twins provide different levels of insight.

Product twins

A digital twin can be created of any physical component, provided you have access to the right data. Product twins are typically easier to create as they represent a single component or product, with the interaction of the components confined within a single assembly. Examples may include a single gear or blade or an assembled rotary motor.

Process twins

Individual product twins can be used to build more complex virtual environments that constitute process twins. The process twin shows multiple product twins and how they interact within
a define process, such as an assembly line within a discrete manufacturing environment.

Plant twins

Product and process level twins can be brought together
to create a virtual model of an entire facility; delivering unprecedented levels of command and control. Accessing
a holistic view at plant level facilitates informed decision
making and allows plant managers to pursue a strategy of continuous improvement.

Project twins

From a macro level, project or performance twins can be
used to monitor enterprise or capital projects from concept
to construction.

Digital twin solutions

The concept of a digital twin should not be seen as a one-size fits all proposition. Although twinning is often spoken of in terms of a single principle, there are many different types of twins, ranging from a single process or asset to an entire organisation.

From single component design to organisational mapping, digital twins typically comprise a physical model, a statistical model and an element of machine learning. For the latter, exception modelling is used to identify, catalogue and manage incidents. For complex twins, such as production lines in manufacturing industries, it is possible to build a number of smaller twins, representing individual assets, and aggregate the individual components into a master model that accurately reflects the entire process.

As technology becomes more intuitive and user-centric, digital twins are commonly represented as 3D models. The logic can be seamlessly paired with modelling technology to render a realistic representation of the process in near real-time. This allows process managers to gain an overall view of the process and better visualise the interactions between individual assets.

As digitalization continues, connectivity becomes ubiquitous and edge computing becomes more powerful, we could see entire facilities managed via digital twins on secure 5G mobile apps.

The applications for digital twin continue to grow and become evermore sophisticated as network bandwidth increases, visualization technology advances and machine learning evolves.

Being able to abstract from reality, or learn from it, opens new avenues of possibility for digital twins. Three areas where digital twin is currently adding value are: