As a follow-on to Tuesday’s blog where I promised to focus on the value of technology to the user and the consumer, today I am exploring the role of digital twins and where they can be used to improve performance and outcomes.
The digital twin is not a new concept. It was introduced by Dr. Michael Grieves at the University of Michigan around 17 years ago. However, recently it has come to the forefront of the dialog as a key component and enabler of digital transformation. This is largely because we now have the wherewithal to collect data at the speed and resolution needed for the digital twin, as well as the deep learning and artificial intelligence algorithms to gain value from that data.
So, backing up, what is a digital twin? Simply put, it is a digital representation of a physical asset or system that can be used to simulate variances. It can be utilized to drive a specific business outcome or to explore specific options.
Digital twins take many forms, from a simple digital twin of a part, which could be used to model the behavior of that part in different environments, to the digital twin of a complex ecosystem that can simulate all kinds of variables to explore the impact of a given change to one or many parts of the ecosystem. In the manufacturing world, the digital twin offers all kinds of options that can benefit the end user, the brand and the manufacturer itself.
The Digital Twin of a Product
Let’s take, for example, the digital twin of a product. This twin is born when the idea is born. The digital thread that runs from ideation to fulfilled solution is, by design, a collection of data. That data, derived from research, from CAD, from the DfM (Design for Manufacturing) process, and from the supply chain and manufacturing process, all contribute to the digital twin.
Once the product is manufactured, it continues to provide data to its twin. If it has sensors, user data can be derived, including environmental data, performance data and of course failure or end of life data. As mentioned before, data for data’s sake serves no one, so where is that data useful? The answer lies in numerous places.
The ability to use the digital twin to model the performance of the next version of the product is one use. Testing that product in new use cases or environments can also be done using the digital twin. If a recall is required, the digital twin can be used to identify the failure and trace it to exactly which group of products must be inspected as well as modeling a repair or upgrade scenario.
The Digital Twin of a Process or Manufacturing Ecosystem
For the manufacturer, the digital twin of the product lives inside and alongside the digital twin of the manufacturing processes and the entire manufacturing ecosystem. Within this digital twin, the manufacturer has a complete granular traceability and a virtual sandbox in which to explore alternative manufacturing scenarios, like moving geographies to deliver on a specific demand, or a sudden ramp in volume. It could also allow for rapid response to disruption in the supply chain. Just about any “what-if” scenario can be modeled using the digital twin, all with no disruption to the “real” processes or factories, and all operating in real-time with live data from the lines and the factory manufacturing execution system (MES), enterprise resource planning (ERP) or product lifecycle management (PLM) systems.
The digital twin drives value for the entire value chain. For the brand and the manufacturer, it delivers visibility and agility. For the consumer, it delivers better value and faster introduction of new products or response to quality audits, since every scenario is tested in the virtual world before reaching the real world.
When it comes to manufacturing in the modern age, it’s all about visibility, agility, and accountability. The digital twin has it all in spades! It is a key catalyst for digital transformation of the manufacturing industry – starting with the digitization of everything!