Understanding Process Digital Twin

Imagine you’ve accepted a lucrative, but unscheduled, order from a prime customer without factoring in a raw material shortage from a critical vendor. Or you didn’t know that your production manager wouldn’t be able to accommodate the order at such short notice. Maybe you forgot that your backup delivery vehicle is currently unavailable. And what if you didn’t anticipate a machine would hit a snag.

Imagine losing money, face and most importantly — a valuable customer.

These examples are not uncommon, and companies have lost millions of dollars in similar situations. Fortunately, we now have a solution that can eliminate such incidents: process digital twin.

What is digital twin? The term (defined as a virtual representation of a physical product) is well-known. It was coined almost two decades ago and has seen various levels of adoption, from remotely monitoring an asset in a manufacturing plant to analyze power use or efficiency, to helping surgeons determine the best surgical procedure for a complex case and ensure the best outcomes. However, it’s only with the recent advances in the internet of things (IoT) that digital twin has become a potential game changer. The market is poised to see “physical” adoption in large scale very soon. Even as I write this, I’m working on several pilots that will help clients realize the promise of this technology.

While digital twin has been adopted for specific devices or products, the real benefits will be revealed when we can adapt and orchestrate it across the entire lifecycle of physical parts, products, systems, enterprises and processes — hence, process digital twin. Here’s an example of how this could work in a manufacturing plant:

• Parts: Adopting digital twin in individual parts can help improve design and predictive maintenance of those parts


• Products: Adopting it across several products can improve engineering, utilization and predictive maintenance of these products


• Systems: Applying it across limited systems can create model-based engineering and help analyze and improve utilization and predictive maintenance of a contained system


• Enterprise: Running it across an enterprise can enhance operations and management support; some call this Industry 4.0


• Processes: Extending it to a process itself can create a truly connected ecosystem and optimize outcomes; welcome to Industry 5.0

We are creating a process digital twin pilot for a manufacturing client that will help deliver the specific outcomes the client is seeking, without incurring any loss of time or money, guaranteeing customer satisfaction. It provides the client with a simulated environment, including all devices, infrastructure and applications, and processes, where they can calibrate, recalibrate and modify any of the elements of the process (such as component availability or variability, machine uptime/downtime, number of staff required or transport uncertainty) for a specific outcome before getting into the real production.

Figure 1: The right side of the diagram shows all the components that feed into the process digital twin, while the left side shows how the different processes, enabled via cloud, connect into the ecosystem to fulfill the process digital twin.

Behind the scenes of process digital twin

With such great potential, what does it take to build such a system? The first tenet, which is also true for any IoT ecosystem, is context. For example, a pharmaceutical company guaranteeing an increase in the production of pills during a potential epidemic is going to be very different from an increase in pill production for general stock-keeping. A process digital twin done right considers both the process/engineering aspect and the industry context.

Figure 2: The design architecture and components for a process digital twin

To ensure success in a process digital twin ecosystem, the system needs to be:

• Modular: The connected ecosystem involves several logical discrete functions; these functions need to be interconnected through a set of well-defined digital interfaces. This ensures modular components that can participate in more than one process to increase productivity.


• Connected: An essential requirement of the process digital twin is connectivity of the logical functions and partner ecosystem via IoT to gain more control, automation and actionable insights. A connected environment requires two-way communication to be able to adapt to the new needs of the connected enterprise.


• Autonomous: To build a simple and scalable system and achieve complex functionality, every logical system should function autonomously and collaborate with other systems to create more value.


• Open:To realize an interoperable connected ecosystem, you need open data formats and the ability to discover data points. Logical autonomous systems further expose these data points through a uniform user interface and enable other systems to retrieve data and take actions.


• Contextual: Contextual artificial intelligence enables the ability to mine data across the process chain and derive actionable insights within the current context. It’s critical, because it enables high-level actions for business improvements and outcomes, which is the key functionality of a process digital twin.


• Secure: The ability to secure end-to-end connected systems is one of the most important and fundamental aspects of the connected ecosystem. Designing for security is crucial with process digital twins; it keeps malicious actions at bay by detecting and potentially denying such actions before they can be performed.

Figure 3: An example of how a connected factory can inform the user, with the greatest accuracy, how much time the order will take to process, based on real-time information from devices

Process digital lets business process innovation and technology innovation join forces to create a man-machine environment that can unlock true value and provide unprecedented transformation for your business. You will then be able to move forward with confidence — growing the business, enhancing the customer experience, improving product quality and increasing cost efficiencies.