The emergence and usefulness of digital continuity, twins and threads, is a direct result of the ongoing process of digitizing the physical world.
Digitization of the Manufacturing World
Digitization is the conversion of the physical world to a digital equivalent. It represents the convergence of the real and the virtual worlds. This conversion has been accelerated with the emergence of the sensor and data rich markets known as the Industrial Internet-of-Things (IIOT). When focusing exclusively on manufacturing and production processes, the IIOT becomes part of the Industry 4.0 evolution.
The traditionally mechanical, pre-digital industry supply chains were very siloed (Refence 1). The evolution to today’s Industry 4.0 required movement beyond siloes to deal with rising complexities and shrinking time-to-markets (TTM). The key to this evolution has been the digitalization of the physical to achieve smart connectivity between things and people throughout the both the design and manufacturing process.
Two consumer sectors still at the early stage of digitalization are the avionics and automotive spaces, in part due to stringent safety, reliability and certification requirements. The avionic – really Aerospace and Defense (A&D) – market is used to long-cycles in both funding and product development. Recent surveys [Reference 2] have shown that only a scant 26% of aerospace companies do business with clients and suppliers in a digital, electronic manner.
Recently, market researcher Accenture has highlighted the drivers behind the digitalization of the A&D sector (Reference 3), which are indicative of other market segments grappling with digitization:
- Flat budgets
- Longer, more complex programs
- A shrinking talent pool of human capital
- Rising demand for products and services
- The explosion of big data, often used for predictive maintenance
- Operational cost optimization due to flat or shrinking budgets
Many of these same factors are drivers in the evolution of the Industrial Internet of Things (IIOT), but none more so than the application of data analytics. This is a natural result of the nature of the IIOT, namely, a system of connected and integrated electronic, electrical and mechanical physical assets that provide raw data for analysis and manufacturing process optimization.
The availability and ever growing amount of this data, in turn, has helped enable the early digital continuity in IIOT market vendors such as GE, Siemens, PTC, CSC, etc. But what does this mean in practical terms?
The Digital Trifecta – Threads, Twins and Continuity
Digitization is the process of converting almost anything into a digital format, e.g., books become e-books, analog music become MP3 bits and bytes, etc. Interestingly, the reverse of digitalization is also occurring as demonstrated by the 3D-printing of a completely digital electronic model.
Digitization is needed to turn a physical system into a digital replica or twin – at least to some degree. This digital representation originated as a by-product of digital manufacturing trend whose purpose was to maintain and re-use digitized production information, e.g., machine settings, specifications, assembly-line configurations, etc. For the manufacturing process, the Digital Twin also incorporates Computer Integrated Manufacturing (CIM), production line equipment/robotics, warehouse and material management. Computer-integrated manufacturing (CIM) refers to the use of computer-controlled machineries and automation systems in manufacturing products that combine both computer-aided design (CAD) and computer-aided manufacturing (CAM) technologies.
Over time and with improved simulation technology, the Digital Twin has expanded to include design activities. Today, Digital Twins comprise a near real-time digital image or software copy of a physical asset or process (see Figure 2). From a design perspective, a Digital Twin is a digital representation of a physical product such as an aircraft engine. Including CAD and related engineering information, it incorporates product specifications, geometry models, material properties and associated simulation information.
Perhaps nowhere is the value of the Digital Twin more evident than in NASA operations. Once deployed, spacecraft are generally inaccessible for repairs. The only way to determine what is wrong with a spacecraft is from information gleaned from sensor systems and transmitted via telemetry technology. When manned missions encounter problems, simulators and Digital Twin databases can help pinpoint the problem, devise possible fixes, and test out repair actions on the ground.
To achieve maximum efficiency, a Digital Twin for the product development, manufacturing and even the entire supply chain will need to be created. This comprehensive goal is still a ways off except in product manufacturing, in which the Digital Twin idea comprises not only the product but also the factory, the equipment and the logistics systems.
Creating a Digital Twin that bridges manufacturing, design and every life cycle phase in-between requires lots of Digital Threads.
The Digital Thread helps extends the Digital Twin into a product’s entire lifecycle, encompassing all data flows across initial architecture, design, engineering, performance, manufacturability and serviceability. It’s a vital thread that runs through all the disciplines, domains and contexts with which a product/service interacts.
The Digital Thread is a framework that enables connected data flows and an integrated view of assets and systems across traditionally siloed elements in manufacturing (and design). The Digital Thread ensures connections between all of a product’s digital assets–and their revisions over the lifecycle –including versions of BOMs, CAM databases, parts, software, electronics, CAD models, documents, requirements, process plans, service manuals, etc
It would be a mistake to imagine that the Digital Thread and twin are similar concepts. The former is data centric path that establishes a connected data flow for all pertinent product data throughout its lifecycle. Conversely, the Digital Twin enables the creation, building and testing of the product in a virtual environment (see Figure 3). By developing Digital Threads, design and product engineers can collaborate with manufacturing engineers to create a virtual, 3D models between the design and manufacturing environments.
If an unbroken, contextually consistent and streamlined flow of data, information and views can be established between the design environment and the manufacturing execution systems, then digital continuity can be achieved. Such digital continuity will allow the information to be updated and constantly available throughout the product’s development lifecycle.
What Twins and Threads are Not!
It is only in recent years that the creation of a truly Digital Twin has been possible. As might be imagined, the digitization of a physical system requires massive amounts of data, computing power, storage, bandwidth and cost. These requirements have been met over the last several years thanks to ever more powerful yet cheaper electronics afforded by Moore’s Law. (3)
A Digital Twin enables companies to understand not only the product as designed but also the system that built the product (manufacturing) and how the product is used in the field (operations and service). Understanding these aspects of the product help companies shrink time-to-market, improve operation, meet stringent safety requirements, reduce defects and more.
Some wonder if these benefits are worth the cost of creating a complete Digital Twin of a new product all at once. How is it even possible to capture every conceivable piece of information about the physical twin that must surely be necessary?
The latter point is an idealization, much like assuming all system-of-system (SOS) projects must start from scratch. Despite the impressive state of today’s computing electronics, no computer system would be able to crunch all the numbers for an exact and complete digitalization of a physical product. Even if all the data could be processed, there would be no way to filter and analyze all the data in a timely manner.
The key to creating a Digital Twin is to start in one area first, presumably the area that is causing problems. As with the engineering of any problem source, the Digital Twin will involve simplifications and assumptions in order to be of value. It will not include every physical aspect of the system, only those aspects that are of interest and value to solve the problem.
To refine our earlier definition, the Digital Twin is really a virtualize representation of all the information needed to supplement existing engineering models and tools to solve a given problem. The creation of the Digital Twin must occur within the scope and context of a given challenge.
Similarly, context is critical for the Digital Thread, too.
Context is Critical
The defining characteristic of the Digital Thread is the continuity of its connection throughout the product lifecycle. Looking at a representation of the Digital Thread (see Figure 4) is very much like looking at a diagram about requirements traceability (see Figure 5). But the Digital Thread is about more than just traceability, it’s about the context and relationship of the connections between all of a product’s digital assets and their revisions – BOMs, parts, software, electronics, CAD and CAM models, documents, requirements, process plans, service manuals, etc.
The relationship aspect of the Digital Thread information addresses both the context and the dependency of the data. For the context, it might answer the question of how a given part is related to another and whether they are both part of the BOM. In terms of dependency, the Digital Thread must reflect how and when data for both parts is changing over time.
This is not to say that the Digital Thread is merely a collection of web links between different data points and subsystems. The threads must be meaningful links between data and subsystems. Nor are this links just a parser-based interconnection of engineering to manufacturing processes or CAD to CAM tools. Instead, Digital Threads must provide the data connections and traceability from concept to end-of-life and across all involved disciplines including software, electronics, hardware, wiring harnesses (for cars and planes), requirements, etc.
In simpler terms, a Digital Thread (there may be more than one) provides traceability to the configuration of the Digital Twin (see Figure 6).
Digitization of Systems Engineering
Some have said that the Digital Thread and its connection to the Digital Twin have put the engineering back into systems engineering. Another way of saying this – at least for the design activity – is that model-based technologies have enabled systems engineering in the digital world. Such digital collaboration enables information (digital) continuity across lifecycle processes (see Figure 7).
If done properly, weaving the Digital Thread will reinforce the basic tenants of system-of-systems (SOS) engineering especially the contextual flow of information. The starting point for the Digital Thread is typically early life-cycle model-based systems engineering (MBSE). This modeling serves as the foundation for all later cross-functional design. As a reminder, model-based systems engineering (MBSE) is the methodology that focuses on creating and integrating domain models as the primary means of information exchange between engineers, rather than on document-based information exchange.
In practice, using the results and insights gained from MBSE and Digital Threads enables early detection of failure modes in product simulations, which in turn lead to less design mistakes. Manufacturing can then link to the resulting Digital Twin to prepare all manner of production assets to build the actual product.
Using MBSE tools, engineers will be able to run system-wide and life-cycle long simulations of products for the Digital Twin to simulate hardware-software plants, products or services at the system level. In so doing, the hardware, software and content functions of the system can be more efficiently managed.
Consider the example of a Digital Twin for a transmission generator as part of a Siemens SIMOTICS general purpose factory motor (see Figure 8). Once available, the motor’s Digital Twin provides up-to-date technical electrical and mechanical specifications, spare parts and operating instructions and more by simply scanning the data matrix code on the physical motor. But what was needed to create the Digital Twin of the generator portion of this motor?
As with any engineering project, one must first determine the requirements before developing an architectural and functional design, performing simulation and implementing the design in a physical system. The basis for this work comes from a variety of contextual sources and is conducted using a multi-tool environment (see Figure 9) that is tied together with at least one a Digital Thread. In the case of a generator, essential design issues would focus on power transmission requirements, voltage loss and static-dynamic system loads.
The result of the associated specifications and analysis work (see Figure 10) is a complete configuration that serves as the Digital Twin. Traceability of the data between the different models and associated engineering disciplines is provided by the Digital Thread. The MBSE ensures a high level of integrated interoperability and digital continuity.
Figure 10: Specifications and design details form the data portion of the Digital Thread which can then be accurately modeled and manufactured.
Many cost, performance and production issues are driving the digitization of the physical world. To ensure digital continuity during this process, Digital Threads must be used to create a Digital Twin of the pertinent portions of the physical entity being designed and manufactured. Further, all of these activities must take place within the context of the system with both the physical and virtual worlds. Understanding the concepts and examples of digitization, Digital Threads, Digital Twins and continuity are needed to develop new products and bring the “engineering” back into systems engineering.
1) Timeline of Industrie 1.0 to Industrie 4.0. Graphic source: Courtesy of DFKI (2011)
2) Stegkemper International Operations Excellence
3) Economist: Cloud Computing Prices Keep Falling
Note: This article was sponsored by Mentor-Siemens.