Digital Thread: from Digital Twins to Predictive Twins and Process Intelligence

The right data at the right place at the right time

Data can turn into information and intelligence or insight depending on how and by whom it is used: data leveraged into intelligence into action in the crossing of digital and physical worlds. The main challenge remains to access and interpret this data or information, at the right place and the right time so that it is can be used as part of new value-added services or products, so that it can help make better informed decisions.

Various data types and models contribute to engineering and manufacturing activities: design models, 3D xCAD models, simulation models, Knowledge-Based Engineering (KBE) models, system-based models, Bill of Materials (BoM) models, mathematical models, cost models, performance models, Bill of Process (PoP) models, collaboration models, etc.

Process models are equality important in terms of operating models and continuous improvement realisation. As data matures throughout the product creation lifecycle, these models evolve and contribute to multiple feedback loops, both vertically (cross-functional) and horizontally (across development and usage phases). These models are also referred as "digital twins" as they are digital "light-weight" representations of the physical products and processes, including the assumptions, boundary conditions and other parameters which link the real and virtual worlds. They allow to run virtual replications of the design and development process, as well as the factory, production, assembly or maintenance processes - enabling two-way connection between the virtual and physical product.

Digital twins as virtual representations of as-designed, as-manufactured or as-maintained products, services and processes (adapted from Grieves, 2014)

Data from physical products (such as data captured from sensors on production machines and products themselves) is used to validate virtual models and their digital twins. In turn, digital twins provide invaluable information to improve and optimise the performance of real world products and services.

Digital twins and the digital thread

While digital twins refer to digital models of a particular asset, the digital thread cut across various functions and data maturity levels, bringing together the data and information flow upstream and downstream of manufacturing activities and related services. As initially defined by Grieves (2014), the digital twin concept is built on three pillars:

1. Physical product in real space
2. Virtual product in virtual space
3. Connection of data and information that ties the virtual and real products together

This is typically achieved by implementing enterprise Product Lifecycle Management (PLM) solutions which must act a Unified Repository (UR) that will link the two products together (physical and virtual).

The digital thread is the creation and use of cross-domain, common digital surrogates of a materiel system to allow dynamic, contemporaneous assessment of the system's current and future capabilities to inform decisions

Connecting the various digital twins provides opportunities for end-to-end business analytics and close feedback loops for data alignment, traceability and optimisation across the digital thread. Overall insights and analytics become exponentially higher, leading to even more possibilities in complex operations.

Digital twins are a dynamic software model of a physical thing or system (Gartner)

Horizontal alignment includes feedback loops, in a 'systems engineering' manner, across design, manufacturing and service phases of the product lifecycle:

• DESIGN: better customer experience, "right first time" design and faster time-to-market
• MANUFACTURING: improved planning, flexible supply chain, leaner production, seamless testing and certification
• SERVICE: predictive maintenance, optimised asset utilisation, smart product upgrade and other value-based services

As Grieves (2014) put it, "this simultaneous view and comparison of the physical and virtual product will reap major benefits, especially in the manufacturing phase of the product." Digital twins provide an evolving digital profile of the historical and current behaviour of a physical object or process that helps optimise business performance. They offer opportunities to adopt a systems engineering approach and focus on two key success factors:

• Enabling smart cost reduction thanks to process right-sizing and increased organisational agility
• Opening the door to alternative business models and are a source for revenue/profit growth

Once the design and processes are virtually validated, the data is transmitted to the factory, where 'intelligent machines' and 'intelligent processes' will translate the data to manufacture the part or product.

What are your thoughts?

References:

• Gartner (2017) Top 10 StrategicTechnology Trends for 2018
• Grieves M (2014) Digital Twin: Manufacturing Excellence through Virtual Factory Replication