Digital Twins Mature Over Time As More Data Is Accumulated

Lionel Grealou CAx, Digital, Enterprise Leave a Comment


Digital twin technology: why and how it matters?

A digital twin is a virtual representation of a physical product or process, used to understand and predict the physical counterpart’s performance characteristics. 

Dr Michael Grieves

Computer aided models are not new. Whatever they are called nowadays, it is fair to state that digital twins have become more elaborated, sophisticated and integrated. Simply put, digital twins are a live digital representations of physical assets (which exist or are yet to be created in the real world).

They are virtual models of the real. Multiple, yet different, virtual / digital twins are used throughout the product lifecycle to simulate, predict, and optimize the product and production system before investing in physical prototypes and assets. 

Digital twins: multi-purpose representations of the real

Digital twins coexist across to product maturity development stages; “all digital twins do not serve the same purpose”, they range from product concept to industrialization and operations. They include CAD and CAE models, but also non-geometrical data and mathematical models. Digital twins also include predictive twins which are to model future state and behavior of a device, product or service.

Delivering value from digital twins, as well as any digital assets for that matter, requires a combination of business change, lean operations, pragmatic delivery and continuous improvement upon service transition from a strategic initiative into business-as-usual maintenance and support.

Knowing that a machine is likely to develop a problem in future is even more important than ‘simple’ simulation lifecycle as it gives the operator time to deal with the problems before they occur.

Digital twins go beyond CAD models and span across the entire “X” lifecycle (where X = product, service, data, etc.)

Digital twin technology moved beyond manufacturing and into the merging worlds of the Internet of Things (IoT), artificial intelligence (AI) and data analytics. As more complex things become connected with the ability to produce data, having a digital hub or equivalent gives data analysts and other professionals the ability to optimize throughput efficiency, assess options and multiple what-if scenarios.

  • CONCEPT twins: aligning start-up virtual twins to carry-over strategy: concepts are architected based on product and platform-based virtual models, including cost models, concept BOM, market and technical attributes, CAD carry-over, make or buy strategy, technology benchmarking, virtual product definition, cost base and virtual build requirements.
  • INDUSTRIALIZATION twins: gearing development digital twins towards seamless delivery: products are developed and industrialized while the enterprise is concurrently getting ready for start of production; aligning product and platform requirements to achieve economies of scale and scope across the extended enterprise and including design and manufacturing partners.
  • OPERATIONS twins: providing in-field operational digital twins for continuous service: from connected products, asset optimization and shelf life extension, on-air  embedded software monitoring and updates, asset performance optimization using machine learning, leveraging IoT, big data and AI to extend the product lifecycle.

Each function, business or technical domain requires different types of models, whose virtual twins are digital representation of the real, tailored for specific context and boundary condition validation. In the concept and industrialization stages, digital twins aim to simulate the physical product or, to be more precise, represent a subset of its behaviour.

Combining digital twins contribute to unify all the data an organization needs to operate and grow.

During the operations phase, digital twins leverage the IoT, product and infrastructure connectivity, support and enhance the way products are used and maintained, continuously embedding new requirements while minimizing disruption and improving customer experience.

Digital twin maturity: from partial to clone and augmented in-field models

The scope of data that is recorded and retained within a digital twin determines what can be known about an asset’s state and condition. The model maturity depends on the data sources, data accuracy, frequency, quantity and quality.

  • A partial digital twin contains enough data sources to create derivative data. 
  • Comparatively, the clone form of a digital twin contains all meaningful and measurable data sources from an asset.
  • Finally, the augmented digital twin enhances the data from the connected asset with derivative data, correlated data from federated sources, and/or intelligence data from analytics and algorithms.

Digital twins can improve and expand over time as more data (and derivative knowledge) is accumulated. Enterprise data model expansion, master data governance strategies and data integration optimization contribute to digital twin maturity improvement.

What are your thoughts?


Disclaimer: articles and thoughts published on v+d do not necessarily represent the views of the company, but solely the views or interpretations of the author(s); reviews, insights and mentions of publications, products, or services do neither constitute endorsement, nor recommendations for purchase or adoption.