Posts Tagged "Digital Twin"

My previous post described the “Digital Twins” in general and the importance of PLM to support it. To begin with a Digital Twin need to provide the means to design, validate and optimize a part, product, manufacturing process or production facility in the virtual world using a set of computer models. It should enable companies to do these things quickly, accurately and as close as possible to the real thing – the physical counterpart. They also need to consume the data from sensors that are installed on physical objects to represent their near real-time status, working condition or position.

Digital Twins was in the making for many years , especially around advanced robotics. Siemens has recognized the value of the digital twin for a long time and enabled the development of full 3D models for automotive body assembly cells. These models were used to simulate, validate and optimize robotic operations before they were executed on the shop floor. With an extremely high degree of fidelity, these applications could not only simulate a cell, but also enable its near perfect virtual commissioning. Advances in computer science have made it possible to broaden the scope of the primitive digital twin to include many more capabilities, information, inputs and outputs. Today Siemens support digital twins for product design, manufacturing  process planning and production using the Smart  Factory loop and via smart products.

One of the most important value of a digital twin is that it enables flexibility in manufacturing and reduces the time needed for product design, manufacturing process and system planning, and production facility design; thus helping companies to develop and introduce new products to the market much faster than ever.  Connecting Engineering , manufacturing process design and actual production is the foundation and starting point for Digital Twins.

A digital twin also improves quality and even supports new business models that offer opportunities for small-to-midsize companies to expand and bring more high-tech capabilities into their shops. Digital twins will help companies become more flexible,  reduce time-to-market and costs, improve quality and increase productivity at all levels of the organization.  When implementing a true “Digital Twin” on the first day becomes a  big ask for companies,  they might want to adopt it in a phased manner, may be in a similar way it evolved – starting with automated manufacturing process design and production.  My next blog will outline the three pillars involved in deploying a digital twin .

Digital twins are the next new thing for product development in this digitalization era. They bring the physical and the digital worlds closer than ever and represent everything in the environment of a physical product, and not just the product itself and its production system.  Enabled by Product Lifecycle Management (PLM), and supported by advanced communications processes and workflows; often described as digital thread, Digital Twins represent the complete physical product throughout the entire lifecycle, end-to-end.

As products become ever more complex due to ever-increasing design complexity, regulatory requirements, higher software content, and the like, conventional simulations can constrain problem solving and decision-making. Digital Twins are much more than the typical CAE simulations with just design specifications,  materials properties, geometric models, components, and analyses such as anticipated behavior under load . It moves past the primary reliance of conventional simulations on geometry. Even the best of today’s simulations are largely limited to geometric data in CAD, CAE, and PDM solutions plus other elements contained in engineering repositories. Conventional simulations are limited to problems that are tightly circumscribed.   Digital twins have no such limitation: geometry and other engineering constraints are just starting points.  Digital Twins are virtual frameworks for managing product data that is orders of magnitude more varied than what conventional simulations handle and more importantly to turn it into actionable information -information that can be used for making decisions and for supporting those decisions as elements of business models.  This new framework uses latest digital technologies to simulate and accurately predict physical product behavior, which can change a business model and provide new revenue and value-producing opportunities; it is the process of moving to a digital business.

The growing importance of digital twins adds to PLM’s key role as the innovation platform. End-to-end digitalization of both products and processes is essential for any enterprise that intends to implement and take advantage of this new models . This means PLM itself must also continually adapt to support the design and delivery of innovative products and services and further enhance its abilities on collaboration, connectivity, and interoperability; which forms the foundations of any innovative platform .

This is a further  followup to my previous articles on digital twins, focusing on the Feedback Loop pillar

Smart Factory loop
The feedback loop starts with the Smart Factory. This is a fully digitalized factory model of a production system connected via sensors, SCADA systems, PLCs or other automation devices to the main product lifecycle management (PLM) data repository. In the Smart Factory, all events on the physical shop floor during production are recorded and directly pushed back to the PLM system or through the cloud. Artificial intelligence (AI) technology is used to study and analyze this information, and the main findings are sent back to either product development
in manufacturing planning or facility planning.
Why is this important? Production facilities and the manufacturing processes tends to change immediately after start of production. New ideas will be implemented, new working methods will be deployed and new suppliers might be selected; all requiring changes to the production system or process. Since these modifications will certainly impact the future, updating them in the system at this stage is becoming a must. Production systems outlive the product lifecycle, and many companies use their production systems to make multiple products. These factors contribute to the increasing need to regularly capture these changes in the PLM system, which can later be used to distribute this information to all parties. The information collected during production can also serve as the basis for improving the maintainability of manufacturing resources. With this information, we can enable much better (sensor) condition-based maintenance, and thus increase uptime and productivity.

Smart product loop
Almost every product made today is a smart product. Many companies are looking for ways to improve the connection with their smart products while they are being used by their customers. Monitoring product use can provide a lot of knowledge for improving products. More than that, connecting to these smart products can generate a new type of business model that may result in more competitive offerings.

PLM Challenge

These feedback loops and the data it generate is a challenge for PLM too. In the short term, the PLM issue for digital twins is how IoT-gathered data can best be put to work—extrapolated, parsed, and redirected? To where? At whose direction? The quick and easy solutions are analytics running on the cloud, machine-to-machine (M2M), and analyses based on Artificial Intelligence (AI). Such questions are expected as digital twins emerge as the next revolution in both data management and lifecycle management.
Ultimately, the use of PLM will allow us to bring digital twins into close correspondence—in sync—with their physical equivalents in the real world. When this comes to pass, we can expect problems to be uncovered more quickly, products to be supported. Products with digital twins will be more reliable with less downtime while operating more efficiently and at lower cost. PLM-powered digital twins will boost user and owner confidence in their physical products. Ultimately, digital twins reflect what users and owners expect to receive when they sign a contract or purchase order.

A classic deployment of a digital twin includes three pillars: product design, manufacturing process planning and feedback loops.

  1. Product design

A digital twin includes all design elements of a product, namely:

  • 3D models using computer-aided design (CAD) systems
  • System models (using systems engineering product development solutions, such as systems-driven product development)
  • Bill-of-materials (BOM)
  • 1D, 2D and 3D analysis models using computer-aided engineering (CAE) systems such as Simcenter™ software
  • Digital software design and testing using application lifecycle management (ALM) systems such as Polarion ALM software
  • Electronics design using systems developed by Mentor Graphics

Using these elements results in a comprehensive computerized model of the product, enabling almost 100 percent of virtual validation and testing of the product under design. All of this eliminates the need for prototypes, reduces the amount of time needed for development, improves the quality of the final manufactured product and enables faster iterations in response to customer feedback.

  1. Manufacturing process planning

The Siemens solutions available today enable the development of three models critical to any manufacturer:

  • Manufacturing process model – the how – resulting in an accurate description of how this product will be produced
  • Production facility model – the where – providing a full digital representation of the production and assembly lines needed to make the product
  • Production facility automation model – Describing how the automation system, including supervisory control and data acquisition (SCADA) systems, programmable logic controller (PLC) hardware and software, human-machine interface (HMI) hardware and software, etc., will support the production system

The value of the digital twin in manufacturing offers a unique opportunity to virtually simulate, validate and optimize the entire production system. It also lets you test how the product, with all its primary parts and subassemblies, will be built using manufacturing processes, production lines and automation.

  1. Feedback loops

When it comes to the feedback loops of the Digital Twins pillars , there are two kinds that have a significant impact on most manufacturers

  • The Smart Factory Loop and
  • The Smart Product Loop.

“Product Design” and “Manufacturing process planning” pillars were in existence for  a while but the “Feedback loops” is a newer one. I will discuss elaborately on it in my next blog .

© Tata Technologies 2009-2015. All rights reserved.