discrete event simulation versus digital twin

Which Simulation Flavour Do You Really Need?

Written by Vincent Béchard on 2023-10-04

Two simulation flavours

Modern simulations got much more attractive and exciting compared to when it started in the 1970s’: high resolution 3D graphics, lots of computational power, data processing capabilities. Simulating a full-scale process plant, mining site, transportation network? Totally reachable to the point where there are so many different simulation techniques and software, it’s getting hard to decide which one to use!

One flavour is traditional simulation, a design sandbox. It is used to model flows, constraints, adapted logic, and usually with 3D graphics. Typical uses: support to engineering studies (validate and review process design), support to operations (find bottlenecks, test improvement solutions). This is a computerized low cost and risk-free testing environment.

Another flavour, which gained in popularity during the recent decade, is the digital twin. It is used to model actual assets and enriched with technical data, maintenance instructions sheets, process centerlines and schedules. It is frequently built starting with a CAD software and BIM (Building Information Modeling) data. A Digital Twin can receive real-time data and emulate the process to detect discrepancies between actual and base conditions. Typical use: support to operations and maintenance. It is a computerized version of “as is” assets.

What both have in common: they look great with modern 3D graphics! Some simulation platforms such as Flexsim also offer truly interactive views, just like Digital Twins do.

The temptation...

person hesitating thinking

Let’s face it: Digital Twins are very trendy! They sit perfectly in the Industry 4.0 ecosystem, they hit imagination and spark high expectations, they really feel modern and futuristic. However, are they the new “best” answer to simulation needs? Asking the question is almost answering it...

By definition, they are the actual (meaning it physically exists!) assets base condition keeper. What is rarely told is the amount of effort required to implement and maintain a Digital Twin. The volume of technical and maintenance documentation, the rate at which data is fed and updated, the software updating process, the interactivity and concurrent multi-users requirements... It doesn’t take long before a Digital Twin database is obsolete! What if an extension to the plant is being studied? How can the twin be used to support the engineering effort, considering that the extension does not exist yet?

Choosing the right simulation approach

Both simulation flavours are useful and powerful analytical tools. Today they are unavoidable valuable support to modern engineering and operations. Let’s stop the war: they are not competing approaches, but rater complementary tools! It’s not one or the other, but it’s one, then the other in an engineering-to-operation lifecycle.

To really get the most value out of a simulation exercise, choosing the best approach must be based on the purpose, not the 3D quality! Here is a quick guide to help selecting the right simulation approach:

  • Do the assets exist?
  • Are the system parameters well-defined?
  • Are trade-off decisions still pending?
  • Are resource counts finalized?
  • Is the layout frozen and the system optimized?

Answering “No” indicates that a Design Sandbox is probably better. Its purpose is to quickly test system or parameters changes and to validate operations feasibility.

  • Is a virtual replicate of the assets needed?
  • Are all the system details known?
  • Is the data pipeline in place and flowing?
  • Is a fast way to reclaim maintenance sheets critical?

Answering “Yes” indicates that a Digital Twin is probably better. Its purpose is to understand deviances from the base conditions and offer speedy ways to intervene.

Asking “Which simulation flavour do I really need” should be easier to answer now!

person feeling relieved

Want to learn more?

At Différence, our core expertise is centered on statistic and data science, Lean applications and operational excellence, and... operations simulation! Don’t hesitate to ask for more information by contacting us.