Digital twins are on everyone’s mind these days. And it’s no wonder, considering the widespread benefits they provide.
These digital representations allow businesses to evaluate the real-time working conditions of physical objects. They enable companies to remotely monitor products and assets – empowering them to quickly detect and fix issues if something goes wrong as well as to make design improvements to future releases.
To follow this up, I recently had the chance to speak with three thought leaders to get their insights on the exceptional value of digital twins, the technology behind them, and what the future holds for businesses that leverage this powerful innovation:
- Monica Schnitger, President and Principal Analyst, Schnitger Corporation (@monica_schnitge)
- Sin Min Yap, VP Strategy, Ansys. (@Sin_Min_Yap)
- Srivathsan Govindarajan, VP, SAP Digital Twin
Which industries will drive the adoption of digital twins?
Monica Schnitger, Schnitger Corporation: Digital twins – and the concept of creating and holding onto digital data about products – are seeing traction in just about every industry. From medical devices, where regulators and consumers want to know what went into making the device and what keeps it operating as specified, to oil and gas, where dangerous and distant facilities make remote monitoring and operations so attractive, digital assets are being recognized as key to efficient operations. These assets form the basis for digital twins, which enable analysis, training, and many new industry-specific business opportunities.
Sin Min Yap, ANSYS: Many companies have challenges in maintaining assets that are typically in remote and hostile locations, where service operators’ health and safety is paramount. So it’s no surprise that companies in mining, oil and gas, renewable energy, as well as construction sites with heavy machinery are early adopters in the Internet of Things (IoT) and digital twin technology.
Srivathsan Govindarajan, SAP: Digital twin adoption is still in its early stages. Currently, we’re seeing a lot of interest in implementing this technology for expensive industrial equipment in industries such as renewables and oil and gas. However, as the technology matures and scales, we can expect almost all assets to have a connected digital twin that is aware of the object’s current state and also of all historical data.
Which use cases can deliver the most value?
Srivathsan Govindarajan: Digital twins can be used as feedback mechanisms in the design process to accurately understand how operational assets age. This can significantly improve the design of assets. Going forward, it will be possible to test a new version of a design by simulating the asset under the real physical conditions that were experienced historically by a previous version of an operational asset. This can help to measure and quantify outcomes of a redesign as well as prevent design errors.
Monica Schnitger: Manufacturers can optimize a production line by considering operating data from each major piece of equipment on the line, which could give insight into its predicted life and deteriorating efficiency over time to forecast what will happen next. Once that data is in hand, the operator can simulate the economic trade-offs between halting production to perform maintenance and what might happen if production continues. Many operators perform reactive maintenance (“Something is about to fail. What do we do?”) or time-based maintenance (“It’s Tuesday, so we add coolant to these five machines”). Predictive maintenance will enable operators to plan much more effectively. This leads to more reliable operations, improved safety, higher quality, and greater profit.
Sin Min Yap: Digital twins can unlock tremendous energy/fuel savings. Take a wind turbine for example. IoT enables companies to leverage multiple inputs, such as weather and wind predictions, expected demand for electricity, etc., to make decisions ahead of time about how best to run which turbine in the system. Analytics and optimization offered by IoT can reduce fuel consumption in power plants by tens of thousands of tons of coal every year while still maintaining the same megawatt output. Companies can also shift from regular, time-based maintenance and reactive equipment repair to predictive maintenance. By leveraging predictive analytics, companies can maximize equipment uptime and increase productivity.
What technologies enable digital twins?
Srivathsan Govindarajan: Key enablers of digital twins are engineering models – such as CAD and Engineering Simulation – cloud computing, IoT, Big Data, and enterprise software that connects the end-to-end value chain from product design through asset management.
Sin Min Yap: Some of the key technologies include analytics; machine learning; virtual, augmented, and mixed reality; enterprise resource planning; product lifecycle management; supply chain management; and systems modeling and simulation.
Monica Schnitger: We’ve had the basics for digital twins for many years. CAD and engineering simulation models of the physical object were likely used in designing modern industrial equipment. Manufacturers have been gathering real-time operating data for decades over internal networks for storage and after-the-fact analysis. What’s new today is the ability to quickly combine the models and data for in-stream analysis. We’re now able to move relevant sensor readings from machine to analytical engines, and we have enough processing horsepower to analyze that data fast enough to inform real-time decision making. We’re at one of those moments in time where technologies are converging to make this possible.
How will the use of digital twins evolve in the next five years?
Sin Min Yap: Customer journeys will be driven by the verticals they’re in, the assets they have, and the age of their equipment. IoT and digital twin technologies have to advance to a point where they become pure decision-making tools. Analytics, product lifecycle management, and optimization algorithms must become black boxes. Customers need to experience the impact before they adopt. They will mature along the same trajectory as analytics – from simple, actionable insights to complex ones.
Monica Schnitger: In the next few years, we’ll see commercial, off-the-shelf applications that will collect sensor data, process some of it locally, and send relevant data on to a central, more powerful analytical tool. This will enable local, immediate action, as well as insights across plant sites or events.
Srivathsan Govindarajan: Currently, digital twins are targeting some very early use cases where they can demonstrate business value by monitoring high-value assets. In the mid-term, as more advanced edge technologies emerge, digital twin precision will increase. By accurately mirroring the physical state of an object – and leveraging machine-learning algorithms – the digital twin will begin to make self-diagnosing, self-repairing, and self-regulating systems a reality. This opens the path to higher efficiency, improved safety, reduced downtime, and lower costs. We can also expect to see artificial intelligence (AI) and machine learning combining with simulation-based digital twins to create some interesting possibilities with respect to predictions that are hard to pinpoint right now. But this combination could really create the next level of insights.
As we have learned from Sin, Monica, and Srivathsan, there’s no better time than the present to embrace digital twins. Digital twins offer companies a groundbreaking new way to generate unprecedented business value.
As organizations discover the significant impact digital twins can have on their enterprises, companies will run to adopt the game-changing technology. But what’s fresh and innovative today will become commonplace tomorrow. So don’t delay your digital twin journey any longer.
To learn more, register for the Digital Twin info brief.
Follow me @howellsrichard.