Digital Twin Integration in CAD: Shaping the Future of Design

There are several ways of integrating the real world into the digital world. These include digital twins, virtual reality (VR), augmented reality (AR), and mixed reality (MR). However, these methods are implemented differently, albeit tied by a common thread: computer-aided design. We have previously discussed computer-aided design (CAD) in VR and the integration of virtual prototyping, a product of CAD/CAE/CAM systems, with technologies like AR and MR. This brings us to the concept of digital twins. What is a digital twin? How does it relate to and integrate with CAD? How are digital twins in CAD shaping the future of design? This article aims to answer these questions. Let’s get started. 

Exploring the Concept of Digital Twins

PTC’s Illustration of the Digital Twin of an Airplane Engine (source)

A digital twin is an up-to-date virtual (computer-generated) representation or replica of an operational process or task, real-world physical product, place, or person. To boost the accuracy of the representation, digital twins heavily rely on data. This data tells the story of the product, process, person, or environment. The data is directly sourced from the physical object or environment it represents and is used to build and subsequently improve the digital twin. Thus, the digital twin is uniquely tailored to only what it represents; after all, no two products, tasks, individuals, or processes are alike.

Usually, this data covers all project stages, from planning and design to manufacturing/construction and use. It is provided by Internet of Things (IoT) sensors and cameras as well as the professionals working on the project. This means the data is not only up-to-date and delivered in real time, but it is also accurate. It therefore provides the digital twins with reliable information that professionals can trust to mirror the exact state of the physical world. It is worth pointing out that this data is the crucial difference between a digital twin and a regular CAD model or physical-based simulation.

Broadly speaking, a digital twin acts as a singular source of information about a project, helping improve collaboration. Moreover, it provides all stakeholders with more profound insights into products, processes, environments, and personnel involved in the project. It is also worth pointing out that multiple digital twins can be integrated, providing a more enriched understanding of the interdependencies and the ecosystem within which they exist.

Building Blocks of Digital Twins

We have established that for a digital twin to be termed as such, it must have real-time or near-real-time data associated with it. But data is only one element that contributes to the wholeness of a digital twin. What we mean by this is that there are several building blocks that provide/gather, visualize, process, transfer, analyze, or use this data. The building blocks of a digital twin include: 

• Modeling
• Sensors
• IoT connectivity
• Compute

Modeling

The modeling building block includes both visualization and scientific modeling techniques. Visualization-based modeling involves the creation of 2D drawings and 3D models using CAD and Building Information Modeling (BIM) software. More on this below. On the other hand, scientific modeling techniques aid in predictions and what-if scenario planning. They predict the behavior of the digital twin and, by extension, the physical object. They are concerned with phenomena like structural deformation, fluid flow, biochemical processes, and more. The predictions are based on artificial intelligence (AI) and machine learning technologies. 

Sensors

Sensors are mounted on the physical object. They collect in real time the data relating to the behavior and characteristics of the object. The digital twin uses this data to update itself. Sensors enable professionals and building owners to identify areas where the structures are aging or nearing failure.

IoT Connectivity

IoT connectivity ensures a two-way flow of data to and from the digital twin. This building block ensures the virtual replica is a ‘living’ and ‘breathing’ representation of the physical object. Nonetheless, it is worth noting that the frequency with which the data should flow from the physical object to the virtual model varies based on the use cases. Some use cases require real-time data transfers, while others thrive on periodic data transfers.

Compute

The last building block of digital twins is computing power or compute. It is concerned with analyzing large volumes of data, helping make sense of whatever the sensors gather. The compute processes the data, converting it into usable insights. It is worth noting that the computing power should be tailored to the scale of the project and the number of sensors to reduce bottlenecks that may restrict the efficiency of a digital twin. At-scale compute is usually the reserve of cloud computing, which also efficiently stores the data. 

Linking these building blocks is a communication framework called the digital thread. The digital thread facilitates the flow of data between the real-world asset and its digital replica, with this data flow occurring across multiple interconnected informational nodes (data sources) and data storage units. It is not only integral to the creation, maintenance, update, and utilization of the digital twin but also ensures and supports the real-time, accurate evolution of the digital replica.

History of Digital Twins

The idea of using replicas to represent actual products is not new. The National Aeronautics and Space Administration (NASA) pioneered the concept in the 1960s. While it was not called the digital twin and was, in fact, not digitized, it had all the hallmarks of the technology as we know it today. At that time, NASA created, at ground level, physical duplicates of spacecraft, notably used during the Apollo 13 mission. The employees who worked as flight crew used these replicas for training and simulation. 

Technologies evolved between the 1960s and the end of the 20th century, with advances in CAD and evolutions in CAM being prominent. So evolved were the technologies that it was possible to depict digitally a physical object. Scholars like David Gelernter, who in 1991 published a book titled “Mirror World,” began imagining a future where people could start seeing the representations of realities through a computer screen. Later that decade, in 1998, the term’ digital twin’ was used for the first time to refer to the digital version of a physical object. 

However, it was not until 2002 that Dr. Michael Grieves, a professor at the University of Michigan, introduced the “Conceptual Ideal of PLM,” which had all the characteristics of the digital twin model. For the avoidance of doubt, it was not called the digital twin, at least not yet. The credit for coining the name for the digital twin model Grieves had introduced goes to NASA’s John Vickers. The term’ digital twin’ – and the modern meaning attached to it – was officially coined in 2010.

Types of Digital Twins

There are four types of digital twins:

1. Product: This type represents the product lifecycle from conception and design to manufacturing, customer use, and decommissioning.
2. Process: This digital twin represents production operations, manufacturing activities, and all tasks used to create products and services.
3. People: A digital twin of people captures data about individuals such as workers or patients. Organizations also use this type of digital twin to provide workers with information about the tasks they are required to complete. By capturing and delivering this data, this type of digital twin helps improve the efficiency of various processes in the product lifecycle.
4. Spatial: This digital twin represents places and physical environments such as workstations or factories. It helps professionals to visualize environments. As a result, they gain more information and insights into the complexities of these environments and how to better engage with them.

How Digital Twins Add Value to Physical Assets

Digital twins are not just intended to represent processes, people, products, or places. Instead, they are designed to add value to each of these assets. It is worth pointing out that the value companies draw from digital twins depends on how they intend to apply the technology. So, how do digital twins create value?

1. Digital twin CAD helps professionals and owners visualize their physical assets, facilitating easy access to and interpretation of data; this is particularly beneficial when the assets are in remote locations or are dangerous, e.g., nuclear power plants, aircraft engines, and machines.
2. Digital twins with simulation technologies can analyze various outcomes
3. Digital twins use measured and derived data from sensors to diagnose/identify and troubleshoot problems; they can suggest the cause of certain issues, enabling staff to deal with them quickly before they escalate. The data also helps managers and owners view areas where the building or product is failing or aging.
4. A digital twin can predict the future state of a physical product or environment, including its future performance, potential issues that may befall it, and the best time to carry out maintenance works.

Benefits of Digital Twins

There are numerous benefits of digital twins, including:

1. Digital twins lead to better product quality thanks to the large volumes of data that engineers can employ when creating future products.
2. They improve operational efficiency by identifying issues and providing potential solutions.
3. They increase customer satisfaction due to predictive maintenance because the digital twins inform suppliers of machines when to carry out maintenance, helping their customers avoid unplanned downtime. 
4. Digital twins reduce time to market by enabling companies to iterate and innovate faster and with greater efficiency
5. Digital twins improve productivity due to less downtime
6. They enhance supply chain resilience and agility, leading to on-time delivery
7. Digital twins lead to better planning, design, and construction
8. They improve the efficiency of products, buildings, processes, and environments
9. Digital twins reduce development costs by facilitating tests that help product designers identify clashes between components and simulate different environments. (The digital twin technology can be combined with virtual prototyping for better results.) 
10. Digital twins enable collaboration between multidisciplinary and cross-functional teams.
11. They guide companies in decommissioning equipment that has reached its end-of-life. Specifically, digital twins provide data that inform decisions such as reconditioning, reusing, recycling, or scrapping the equipment.

Digital Twins and CAD: A Symbiotic Relationship

Digital twins are nothing without the data associated with the products, processes, places, or people they are supposed to represent. They are also nothing without visualization building blocks like 2D drawings, 3D models, and 3D renderings. These visualization building blocks help represent the visual aspects of the data associated with physical objects. Naturally, this is where CAD and product lifecycle management (PLM) come in. 

For a 2D drawing or 3D model to be referred to as a digital twin, it must have data associated with it. Otherwise, it remains a 2D drawing or a 3D model. Gathered by IoT sensors and technologies, this data must be sourced from the physical object it is representing. Simply put, the data converts CAD objects from mere 2D or 3D representations positioned in a virtual space into representations of the physical form and behavior of existing real-world products, environments, tasks, or people. Against this background, all digital twins are CAD models, but not all CAD models are digital twins. 

Technological Backbone: How CAD Supports Digital Twins

CAD is a fundamental building block of digital twins. It enables professionals to visualize the physical object, creating elaborate 3D models and comprehensive 2D drawings. Moreover, CAD tools enable users to create immersive 3D renderings of the physical environment that permit immersive walkthroughs using other technologies like VR, AR, and MR. Modern CAD systems, which ship with built-in computer-aided engineering (CAE) tools, also support simulations. 

Usually, CAD is a foundational step to creating a digital twin; it is the backbone. It is, therefore, accurate to say that there is no digital twin without CAD. In fact, digital twin platforms integrate with CAD and BIM solutions. For instance, Autodesk Tandem, a digital twin platform, is designed to integrate CAD geometry from Revit, geospatial data, facility management data, IoT data, and more. 

Companies have also used several solutions from Siemens Digital Industry Software to create comprehensive digital twins of their projects, with CAD and simulation tools playing a crucial role. Given the need to combine the capabilities of multiple software and tools, selecting tools and software that support interoperability is vital. For its part, Siemens packages the interoperable tools, offering them as Digital Enterprise Services.

Similarly, PTC partnered with Ansys to create a digital twin offering that combines interoperable products from both companies. The offering combines PTC’s ThingWorx, an industrial IoT software, with Ansys Twin Builder, a powerful modeling, simulation, and analysis tool to create virtual replicas of physical assets. While PTC already had IoT software, it had to partner with a company whose solution has the other building blocks of a digital twin, emphasizing CAD’s status as the backbone of digital twins.

Applications of Digital Twins in CAD

You can deploy digital twins in various industries, from logistics, agriculture, and healthcare to manufacturing and architecture, engineering, and construction (AEC).

Digital Twin in Manufacturing

The manufacturing industry has been a hotbed of implementing the digital twin. This is partly because a typical factory already has hundreds or thousands of sensors. Additionally, CAD and CAE use is commonplace during the early stages of product development, with CAM tools also featuring prominently. Thus, digital twins bring together existing technologies as opposed to introducing novel tools. 

Manufacturing organizations use digital twins for predictive maintenance, optimizing the maintenance of machines. They also use digital replicas to observe the behavior and performance of various components that make up the product. This enables them to identify parts that are wearing out faster than anticipated. Armed with this information, companies in the automotive industry, for instance, can arrest this problem early enough, which limits the number of recalls.

Digital twins of the factory floor enable managers and staff to identify issues – and quickly address them – as well as opportunities for improving operations. As a result, digital replicas enable companies to improve delivery and the performance of their facilities. To complement the use of digital twins and their benefits, manufacturing companies are also adopting other strategies.

For instance, some are implementing the model-based enterprise (MBE) strategy. An MBE is a manufacturing company that uses 3D CAD models throughout the product lifecycle to manage business processes. Such a company reaps multiple benefits of the strategy, including improved efficiency, reduced time to market, lower cost, and more.

Digital Twin in AEC

Before the onset and increased proliferation of digital twins in CAD, the AEC industry predominantly used manual workflows and paper-based information exchange. For instance, all data about a construction project or building, as well as its performance over its planned lifetime, was collected physically by personnel who had to be onsite. This means the data was directly and solely housed inside the precincts of the building. The data was then documented using static documentation like computer files and paper. But not anymore.

Digital twins in the AEC industry have obliterated this traditional practice. Now, IoT sensors are installed all through the building or structure, enabling real-time or near-real-time data collection. This means the digital twin is updated regularly to reflect the real-time characteristics of its physical replica. Which breathes life into the digital replica. This has had many advantages.

For instance, in 2021, a digital twin of an eight-decade-old bridge in Norway helped prevent a disaster. IoT sensors installed on the real-world bridge sent notifications of unusual movement – the sensors showed that the end of the bridge was moving up and down whenever a track passed. This led to the realization that there was a problem with one end of the bridge’s support. As a result, officials closed off the bridge and began constructing a new bridge.

Digital Twin in Healthcare

It might be hard to comprehend how the digital twin and CAD technologies are applied in the healthcare industry. But researchers have found that the potential uses of digital twin technology in this industry are limitless. They can be used to monitor the performance of equipment and medical devices. Digital twin and CAD systems can be used to create digital models of health facilities to monitor and analyze care delivery as well as predict the impact of personalizing care. 

Perhaps a more practical use case has been spearheaded by Dassault Systèmes through its Living Heart Project, which began in 2014. The project brought together medical, biomedical, and pharmaceutical experts with a common goal of building and validating a virtual twin of the heart. It is hoped the project will increase industry innovation and wend the way to an efficient pathway for patients to access new treatments for heart disease. The Living Heart Project has inspiredthe Living Brain, Living Lungs, and Living Liver Projects

Another potential application of the technology, which is still under development, is using digital twins of patients. These digital replicas will have the patients’ respective medical histories. Moreover, they are expected to be coupled with IoT sensors that measure and relay patients’ health information for real-time monitoring and evaluation against their medical histories.  

Digital Twin in Logistics

The logistics and shipping industry is quite complex – ensuring products reach their intended destination in good condition and on time is definitely not easy. Moreover, coordinating multiple shipments to and from disparate locations simultaneously adds to the complexity. But one thing is certain: that shipping does generate vast volumes of data that logistics companies can use to ensure faster, more efficient, more eco-friendly, and more secure shipments. That is where digital twin and CAD technologies come in. 

However, in this sector, a single digital twin does not cut it. The companies should integrate multiple digital twins that separately capture information about fulfillment centers, warehouse operations, garages, parking locations, or vehicle locations. Collating the data from these disparate sources can itself be a daunting task. Which is why it is advisable to use at-scale compute based on the cloud.

Digital twins in logistics provide in-depth insights into operations, helping professionals plan, design, and optimize shipping roots and supply chains.

Digital Twin in Agriculture

Digital twin and CAD can be used in the agriculture industry. Here, CAD and GIS systems are used to create digital twins of farms, with satellite imagery capturing data and AI and ML models analyzing it. This data can range from farming activities, weather conditions, and water availability to crop variety and health and soil quality. 

The combination of digital twin and CAD/GIS in agriculture facilitates crop yield and risk predictions. It can enable farmers to automate farming activities like soil preparation, fertilization, and crop rotation. It can also help them predict the planting and harvest times that guarantee high yields. The technology can also be used for monitoring and managing livestock as well as optimizing their population. 

Digital Twin in Mining and Energy Sector

Digital twins enable mining companies to simulate the work environment, equipment, and machinery, allowing the miners to test new methodologies and techniques and create short-term and long-term mining programs. The technology also allows the companies to create estimates of the drilling, crushing, and extraction programs and train their personnel off-site before deployment. The digital twin also facilitates predictive maintenance.

Digital Twin in Infrastructure and Urban Planning

Infrastructure organizations such as rail equipment companies, electricity transmission system operators, and state departments for infrastructure and urban planning use digital twins to simplify equipment maintenance, plan and monitor day-to-day operations, and visualize planned expansions or future projects.

Challenges and Considerations in Implementing Digital Twins in CAD

There are several challenges to implementing digital twins in CAD and, more broadly, various industries. Indeed, real-world products, environments, and processes can be complex, yet their digital twins should capture their inherently complex characteristics and behaviors to a tee. The need for precise digital matching of these physical objects can overshoot the budgetary allocations and available computing resources. It can also exceed in-house data governance capabilities and move away from the organization’s culture. 

In this section, we will discuss the challenges companies and professionals face when applying digital twins in CAD. We will also discuss the considerations they can make to limit the impacts of the challenges.

1. Prohibitive Cost

Implementing digital twins in an organization requires substantial investments in CAD software, sensors, cloud computing infrastructure, IoT dashboards, and other supporting technologies. The investment can cover 3D model and AI model development. The cost can be prohibitive, preventing companies from realizing the true potential of the digital twins. 

Fortunately, you do not have to deplete your bank account chasing a complete digital twin product. First, comparing the digital twin approach to alternative, more pocket-friendly approaches is essential. You might be shocked to discover cheaper alternatives that can deliver the same value as more expensive digital twins. You can also opt out of picking a cloud computing provider and instead use a conventional database. 

Nonetheless, the cost of technologies that enable and enhance the digital twin is dropping. This has accelerated the adoption of digital twin, CAD, and associated technologies.

2. Poor Data Quality

A good and reliable digital twin is associated with equally reliable and quality data. However, that is not always possible with large-scale projects. These projects are characterized by hundreds or thousands of sensors that operate in ever-changing and demanding field environments and communicate over flaky networks. Unfortunately, these factors work in concert to lower the data quality, limiting the creation of good digital twins.

To get around this problem, companies will need to come up with methodologies to identify and isolate the poor-quality data. Moreover, they will have to find ways to bridge the information gaps and inconsistencies arising from these measures.

3. Imprecise Representation

A digital twin should replicate its physical counterpart exactly. However, it is not feasibly possible, at least presently, to match the thermal, electric, chemical, and physical properties and characteristics of physical objects. And in cases where it is possible, the process is costly, time-consuming, and challenging. This challenge forces engineers and designers to simplify their creations and make assumptions. These considerations enable them to find a middle ground where there is an ideal balance between the desired characteristics of the twin and the cost-related and technical constraints.

4. Data Security and Intellectual Property (IP) Protection

As we have repeatedly mentioned, a digital twin is a creation of data. Usually, this data is proprietary. It can relate to patented or copyrighted products or product designs that are considered trade secrets. Moreover, the digital twin may contain sensitive data relating to usage and customer processes. 

The sheer volume of data required to create a digital twin gives rise to challenges around data security and IP protection. In this regard, companies should use up-to-date measures to secure their data to prevent costly data breaches.

5. Cybersecurity Threats

The various building blocks of digital twins form a large attack surface that cybercriminals can target. And given the need for hundreds or thousands of internet-connected sensors, the attack surface grows even larger. Of course, criminals’ intentions can be driven by the importance of digital twins and their associated data to organizations’ operations. By accessing proprietary data, criminals may want to extort companies into giving them large sums of cash so they do not release their illegally obtained find or compromise the digital twin. 

The various unwanted outcomes of security breaches point to the need to prioritize cybersecurity. But the effective management of the cybersecurity of digital twins may not be easy for some organizations given the sheer number of software, parts, and professionals needed to create a working digital twin.

6. Difficult Change Management and Capacity Building

Whenever management opts to introduce digital twin technology to an organization that has previously not implemented it, there is bound to be resistance or sluggish adoption. Change is hard, and when the change involves as complex a technology as the digital twin, then it becomes even harder. This presents a critical challenge around change management and capacity building. 

Organizations must make sure their employees possess the necessary skills to work on digital twins. Moreover, companies must find ways to motivate their employees sufficiently to make the shift. While these requirements are not easy to achieve because of the need to profoundly shift the organizational culture, companies can turn to education to solve this challenge, seeking the instructional guidance of experts in digital twins and technological transition.

7. Lack of Interoperability and Data Standardization

A digital twin is a summation of numerous parts, most of which are provided by different vendors and suppliers. While they are supposed to work in congruence, at least in theory, this is not the case in practice. Moreover, AI and simulation tools, which are expected to all serve the same function, may not support this capability. It is not uncommon to find an AI tool or simulation application from one provider that is incapable of replicating the capabilities of another supplier’s product. This creates the challenge of interoperability.

Another challenge is the lack of data standardization. This challenge prevents data from one building block of a data twin from being integrated into another. For instance, data from a digital twin CAD software may be stored in a proprietary file format that cannot be opened using any other tool. To get around this problem, organizations should use standardized file formats and look for tools that support interoperability.

Conclusion

Digital twins are increasingly becoming commonplace in multiple industries, shaping the future. The growing adoption of digital twin technology is driven by the convergence and increased evolution of technologies like the IoT, sensors, artificial intelligence, machine learning, cloud computing, simulations, and more. The technology is using data to breathe life into the once-static CAD models. With it, many benefits abound. 

Digital twins can describe physical assets, diagnose problems, analyze outcomes, and predict future events. This value has seen companies and organizations in the manufacturing, healthcare, mining, infrastructure and planning, agriculture, and logistics industries adopt digital twins. However, their implementation has not been without a fair share of challenges. From poor data quality and lack of data standardization to difficult change management, cybersecurity threats, imprecise representation, and the need for IP protection. Fortunately, there are ways to get around this problem.