10 reasons to combine digital twins and synthetic data

Hear from the CIO, CTO and other C-level and senior executives on data and AI strategies at the Future of Work Summit on January 12, 2022. Learn more


Artificial data and digital twins are complementary approaches to refining on real-world data to improve AI and product design. Synthetic data tools generate labeled data from a small subset of actual data to train AI. Digital twins generate “what-if” scenarios to evaluate various performance, cost, and sustainability trade-offs.

Digital Twins can help expand synthetic data tools to support real-world digital transformation in construction, medicine and the supply chain. In contrast, synthetic data can help teams using digital twins to simulate different scenarios more efficiently.

These tools focus on different markets and use cases for that time. Synthetic data tools focus on improving the AI ​​development workflow. The capabilities of digital twins are added to industry-specific applications for product development, manufacturing, construction and medicine.

“Synthetic data and digital twins based on real-world data will co-exist and complement each other in specific cases,” Gaurav Gupta, partner and global head of digital engineering for technology research and advisory firm ISG, told VentureBeat. Synthetic data will be the preferred choice when price is limited due to logistics or privacy reasons or when actual data is unpredictable or unavailable. Digital Twins will be preferred for applications such as predictive maintenance that require a closed loop between the product-in-use and its digital equivalent.

But the two can also complement each other, he added. Synthetic data can enhance the digital twin application, for example, where the model needs to be shared with various stakeholders, but without any of its private, sensitive or classified information. In such cases, the synthetic data model may act as a proxy for the digital twin models. Similarly, the artificial data model, by contrast, is supposed to be, by definition, as close as possible to the real-world view. As such, the existing digital twin model can feed into or accelerate the synthetic model design.

Here are 10 ways to complement synthetic data and digital twin capabilities in practice:

1. Improving decision making

Gill Albaz, pioneer of simulated data for human-centered AI, chief technology officer and co-founder of Datagen, told VentureBeat that synthetic data tools can report three components that support better digital twins for decision-making. Synthetic Motion Generation component designers can help people explore how it will work before it becomes a crowded wing of a plane, a busy alley, a new bike lane or a store. The synthetic visual generation component can help designers explore the visual parameters of the digital double and see how the new building looks at different times of the day, in different colors and from other convenient points. The synthetic physics modeling component can help designers see how the sun falls on the roof, heating certain areas and powering solar panels. The next step is to simulate the cell tower reception, the effects of earthquakes and floods.

2. Urban planning

Kevin Saito, senior product manager at Unity, told VentureBeat that the artificially created environment could allow architects to understand what would affect architects to more appropriately place windows on their facilities or increase planning considerations. For example, Vu.City in the UK helps architects and city planners understand the three-dimensional aspects of a project in a fully integrated urban landscape.

Saito said use cases drive different priorities. Digital twin simulation requires something that is going to interact with the visual, making physical properties important and visual fidelity, and as a result, low visual fidelity can be preferred because it is cheaper to generate and run on scale. . What is important for synthetic data generation is that it matches the visual realism of the visual environment to mimic it. If you are generating synthetic data to train a computer vision model, visual fidelity is important, but you do not need to mimic the actual physical properties of the environment. For example, physical properties such as weight, center of gravity, or friction that have no visual features are unnecessary.

3. Create new scenarios

The Digital Twin Leader PTC defines the digital twin as a data-driven representation of specific physical machines, people, and processes, rather than a simple process or digital modeling of a machine. “The concepts of synthetic data and the digital twin are closely linked,” Ed Cuoko, vice president of analytics at PTC, told VentureBeat.

Just as physical machines and processes produce data to represent operations, digital twins produce synthetic data to represent simulations of those operations. As such, the scenario generation is one of the primary use cases of the digital twin. “You could argue that visual creation and resolution would almost certainly be a universal use case for the digital twin as it evolves,” Kuoko said.

4. Individualize the medication

Synthetic data can help overcome privacy challenges when using digital twins to improve healthcare. Unlike airplanes or digital twins in cities, precision medicine is more difficult to define the behavior of such models and requires access to highly controlled healthcare data. “High-quality, synthetic healthcare data is emerging as a mechanism that enables digital twins in precision medicine to become a reality, without significant privacy or governance barriers, in collecting and merging such large datasets,” Syntagra co-founder and chief technology officer said. Offer. Mendeleevich told VentureBeat,

5. Validate medical models

Ben Alesdorf, a consultant at TLGG Management Consultancy, told VentureBeat that synthetic data tools and digital twins could be combined to generate test data sets that could expand and validate healthcare models. For example, synthetic data can help confirm that a model developed for a specific population or demographic retains accuracy when applied to another population. Cases of actual use in the medium term, ”he said.

6. Surface supply chain problems

Suketu Gandhi, a digital supply chain partner and global leader at Kirney, a strategy and management consulting firm, says supply chain digital twins can plan around novel supply chain tensions. While real-world data often maintains a steady rhythmic motion, synthetic data tools can insert large waves into the data.

“The key is to understand how to create truly random synthetic data, such as a dramatic change in channel buying behavior by consumers or a 400 per cent increase in demand, to understand vulnerabilities in the supply chain or customer service,” Gandhi said. . All of these can help assess the impact of extreme conditions on the resilience of systems. Also, running thousands of scenarios can help bring out new product opportunities, competitive threats, and opportunities that are not visible with real-world data, such as unconventional partnerships.

7. Simulate failure on a scale

Veritas combines synthetic data and digital twins to improve data security. Veritas utilizes digital twins built on many years of telemetry data from more than 15,000 Veritas netbackup appliances. Both digital twins and synthetic data help AI engines predict before failures and provide smart predictions, enabling customers to better plan their structural needs but also provide information about the reliability of their estate. doing.

Eric Seedman, senior director of Veritas Technologies, told VentureBeat, “The combination of synthetic data with digital twins helps our overall AI models interpret the factors affecting system performance and data reliability by detecting discrepancies in both devices and data.” The combination system helps to ensure uptime and identify changes in data and parameters that may indicate malware or ransomware infiltration.

8. 5G upgrade

Telcos is exploring how to combine digital twins and synthetic data to evaluate diversity in equipment, placement and protocols for the new 5G deployment. Stephen Douglas, head of 5G strategy at Sprint Communications, said digital twins are currently being used to validate new 5G infrastructure before launch. These tools mimic the network components for a new network component or system under test.

Digital twins use synthetic traffic to simulate complex validation scenarios for the component under test. This synthetic traffic is based on real-world traffic and is driven back by the digital twin to see how the component copes with stress. They can also simulate vulnerabilities and cyber attacks and create corner cases that are difficult to replicate in real-world 5G networks. It expects similar capabilities to improve operational networks in the future, so that the best instrument settings can be identified, validated and recommended in real-time.

9. Look for unexpected failure modes

Douglas said the combination of synthetic data and digital twins could also improve performance and load testing for connected vehicles to help engineers understand how their systems will cope in future circumstances. For example, a connected autonomous vehicle may experience poor connectivity due to congestion in a congested inner city, or the signal may be weakened or lost. Synthetic data variations and digital twin emulated replicas can help eliminate unforeseen problems to create more resilient systems.

10. Improve customer experience

Synthetic data can help extend customer experience data to digital twins of customer experience to improve product design, cost or sales. “We frequently use this in enterprise software that converts customer data into synthetic data which is then used to test the digital twin,” Vince Padua, CTO of Xway, told VentureBeat.

These digital twins reflect the actual customer usage data of enterprise software products such as whether customers use a particular feature, how they decide to receive notifications from the product or how they interact with other users while using the product. This usage data can be collected, anonymized and synthesized to drive test automation, improve product roadmaps and increase overall customer satisfaction. At some point, usage data can identify patterns that can be automated with AI or create a ‘digital twin’ of the customer experience in which AI can be given tasks to determine the fastest way to solve them.

Venturebeat

VentureBeat’s mission is to become a digital town square for technical decision makers to gain knowledge about transformative technology and practices. Our site delivers essential information on data technologies and strategies so you can lead your organizations. We invite you to access, to become a member of our community:

  • Up-to-date information on topics of interest to you
  • Our newsletters
  • Gated idea-leader content and discounted access to our precious events, such as Transform 2021: Learn more
  • Networking features and more

Become a member

Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *