Digital Twins in Manufacturing: Producing Data, Not Just Finished Parts or Products

Digital Twins in Manufacturing: Producing Data, Not Just Finished Parts or Products

By Jason Busch and Lisa Reisman

Over dinner with two of our friends on Saturday night — one, the head of IT and digital for a manufacturer in Chicago, the other a doctor — we somehow got onto the subject of digital twins. And no, we’re not talking about some type of online replica or “bot” that mirrors the intelligence or better behavioral elements of a real person. The concept of a digital twin is to have an actual, digital version of a real-world actual object – that reports on and represents the activity of the part, component, finished product or asset (e.g., stamping press) that it represents.

Wikipedia defines a digital twin as a “a digital replica of physical assets (physical twin), processes, people, places, systems and devices that can be used for various purposes. The digital representation provides both the elements and the dynamics of how an Internet of things device operates and lives throughout its life cycle.”

From a pictorial or graphical perspective to understand a digital twin, imagine a 3D representation of an actual robotic assembly arm. The twin, which can be called up at any time on a desktop screen, tablet or phone provides an outside-in – or inside-out – view of all of the reported data that is coming back from the asset. But what would you do with this data?

Sight Machine, a digital twin software provider, presents a number of use cases. According to Spend Matters, an industry research publication, Sight Machine offers “a ‘digital manufacturing platform’ that mirrors the manufacturing and production process (building a “digital twin”) to provide enterprise manufacturing visibility across all plants and suppliers and analytics on top of the exposed data. The analytics workbench allows for root-cause identification (when something breaks), plant and production optimization, statistical process control, anomaly detection, KPI calculation, and predictive (trend) models.”

In other words, Sight Machines uses digital twin technology to capture signals and data that represent the actual manufacturing asset itself – which can provide insight not only into the underlying root cause of an issue, but also general pattern recognition and anomaly detection on a 24/7 basis. Sight Machines is one of hundreds of different technology providers (including stalwarts such as Oracle and SAP, not to mention dozens of start-ups) which are investing in digital twin and sensor-based capabilities.

As digital twin technology begins to ship as a component of assets outside the production environment itself, we believe this technology could prove more transformative to the world of production and aftermarket support / maintenance than just about any type of disruptive capability. When combined with other technologies such as self-learning artificial intelligence systems that can teach themselves – after being taught on training data sets – to spot patterns that even a human can miss, digital twins could pave the way to help avoid disasters such as the recent Boeing 737 MAX tragedies by predicting mishaps before they occur.

But a digital twin cannot exist in a vacuum – it needs data to feed it. Fortunately, the cost of sensor-based technology, including the type of remote units that connect just about anything to the Internet (e.g., a sensor within the engine of a CAT piece of equipment which senses an imminent failure of a core system unless virtual or physical maintenance is initiated) continues to drop. In 2008, a typical IoT sensor cost nearly a dollar – that number dropped by over 50% in 2018 to 44 cents, according to research site Statista. And it promises to continue to decline.

Digital twins bring the promise of a safer and more reliable production world – and safer and more reliable products. While technology to support the mirroring of both production and real-world usage environments exists today, we expect the use of this capability to grow significantly in the years to come.

Jason Busch and Lisa Reisman are Editors at Large.

Surplus Record Index of Available Machinery & Equipment

Comments are closed.