Intelligent digital twins and the development and management of complex systems
The interest in defining and implementing Digital Twins (DT) is at the forefront of today’s product organizations. The evolution of increasingly complex systems requires that their information be organized and managed via digital twins. In addition, the development of Artificial Intelligence (AI) will result in an increasing degree of system complexity. These complex systems will exhibit true emergent behavior as AIs modify system aspects as theresult of goal seeking and learning. DTs will need to increase in capability becoming intelligent in their approach. This article presents a discussion onhow these Intelligent Digital Twins (IDTs) will evolve and assist in developing and managing complex systems.
Why it’s relevant to Nextspace
This article is of interest to Nextspace Partners because of Michael Grieves’ review of the Nextspace platform, after which he commented:
I like Nextspace a lot from an architectural standpoint. The platform is positioned to address digital twin interoperability and is also indicative that digital twins are evolving into the next stage—platforms. Dr Michael Grieves Chief Data Scientist at the Digital Twin Institute
The article is a good length and worth reading in full, but for those pressed for time here’s a summary:
Grieves discusses the evolution of digital twins and his role in it. He then begins to outline the journey of creating a digital twin – from prototypeto launch to multiple twin aggregation.
He breaks this journey down into several phases based on the product lifecycle, briefly outlining the key characteristics of each phases:
- Create
- Build
- Operations
- Disposal
Whilst he breaks the twin development into phases to keep it product-centric, Grieves strongly argues against functional siloing creating a twin for engineering and another twin for manufacturing.
Grieves states his vision clearly:
“The ideal future state is that we would like to design the product, test the product, manufacture the product, and support and sustain the product all virtually.”
An observation made by Grieves is particularly useful for Nextspace Partners:
“Even though the create phase may be short in comparison to the entire lifecycle of a product that may span decades, decisions in this phase have a major impact in determining future product costs. Estimates of product cost determination during the create phase are as high as 80%. There is an increasing ability to perform virtual testing at a fraction of the cost and in less time to replace physical testing. This has the potential to reduce costs, improve quality, and reduce time to market.
Grieves then discusses the ability for digital twin simulations to move us forward in time. He outlines two critical requirements - an understanding of the physics that take place over time and the computing power capacity to run complex simulations.
Next, Grieves discusses simple, complicated and complex systems. A simple system being mechanical cause and effect and a complicated system simply having more mechanical causes and effects. He spends a decent time explaining and separating complex systems as it is complexity that drives the need for a digital twin.
This then leads to a discussion about AI, modelling and simulation providing forms of intelligence for digital twins, and what will happen when these merge.The Intelligent Digital Twin (“IDT”). A transition from passive twins to active twins. Online, goal-seeking and anticipatory.
Grieves goes back through the product lifecycle phases of a digital twin outlining the benefits of an IDT. He differentiates IDTs from Industry 4.0 –giving Industry 4.0 the goal of reducing time from an adverse event to its remediation. Preventative maintenance. Instead, IDTs perform prognostic maintenance based on performance indicators. Predictive maintenance.
Grieves concludes:
Digital twins have emerged as a top technological initiative in the 21st century. The ability to move work from the physical world into the virtual world will enable products to be faster, better, and cheaper6. The proliferation of products that are more complex with emergent capabilities will require the information capabilities of digital twins throughout the entire product lifecycle. Digital twins exist from the moment there is an intended product since the information about a product will always precede its physical manifestation.
Digital twins will need to evolve in order to meet the challenge of increasingly complex products. The advances in AI and M&S will make possible the Intelligent Digital Twin. Digital twins will move from being an information repository to providing constant guidance to their human users.
The characteristics of IDTs are that they are active, online, goal seeking, and anticipatory. This will enable IDTs, enabled by the exponential increases in computing capability, to apply their abilities in moving in time to predict the future by constantly performing simulations of possible futures from current states. The Front Running Simulations (FRS) of IDTs will provide, at a minimum overwatch of their associated Physical Twins and ideally predict adverse condition so as to prevent those adverse events from happening.
The movement of work from the physical world to the virtual world will be a hallmark of the 21st century. AI and M&S will enable Intelligent Digital Twins and will greatly assist humans in the work with developing emergent products. Intelligent Digital Twins will not replace humans.
The article has good references for further reading and was reviewed by 4 peers before being published.
