Having spent five years as a structural analyst in the aerospace industry and a decade as a Mechanica / Creo Simulate instructor, I’ve learned there are two sets of best practices related to simulation and analysis:

• Those related to the functionality inside of the software, commonly referred to as the “picks and clicks.”
• The thinking and work you do before even launching the software.

The latter is far more critical. Let’s discuss some of the best practices related to performing real-time simulation.

Estimate the Answer

Before you jump into Creo Simulate or Creo Simulation Live, you should have an idea of the rough order of magnitude (ROM) of the expected stresses or displacements. This way you will be able to assess whether the results from the software are credible or not. Use free body diagrams, Roark’s Formulas for Stresses and Strains, PTC Mathcad worksheets, and other resources for classical equations and methods for the answers to an idealized (simplified) version of the analysis.

Frame the Problem

In some ways, being a structural or thermal analyst is like being a lawyer. Successful court lawyers aren’t necessarily the ones with the best grasp of case law and precedent. The most successful lawyers are the ones who know how to frame the core issue.

When someone asks me questions about performing an analysis, I ask them what failure mechanism they are worried about. Is it going to deform? Overheat? Resonate? Buckle? Fracture? And where do you think that might happen? Understand the problem you’re trying to solve.

Spend a Lot of Time Thinking About How Your Model Loads and Constraints

You have many choices in how you will represent the real-world conditions that your product will experience. Different loads and constraints will give you different results and minor tweaks to them can significantly impact the results.

As the computer programming saying goes, garbage in, garbage out. It would be great if there was a standard reference manual for how to model loads and constraints for all conditions, but that doesn’t exist. The more thinking you do up front, the better your analysis will be. Try running the model with different approaches. And always ask for help and advice from those who know more than you.

Apply the Appropriate Level of Effort (LOE)

Are you analyzing a model that is human life or flight safety critical? Or are you working on something that could yield and not adversely affect the basic operations of your product? Are your requirements for multiple of factor of safety (FS) or are you right at the margins of safety? The type of problem you are trying to solve determines whether you need the 85% answer, 95% answer, or 99% answer. In my career I’ve seen different levels of rigor for human spaceflight versus disposable $50 computer tablets.

If you’re in this industry long enough, you will undoubtedly encounter the person who never delivers the answer because they can never finish perfecting the model. To paraphrase Voltaire, “perfect is the enemy of good… or the good enough.” Done is better than perfect.

Check Your Answers

Once you start getting values and generating contour plots, take a moment to perform a sanity check. Do the displacements, stresses, temperature, and/or modes make sense to you intuitively? Are they unusually or unexpectedly high or low? How big are those values in the real world? If they’re too big, you may also need to consider non-linear analysis.

This video shows some best practices related to the functionality / “picks and clicks” side of real-time simulation in conjunction with the design process:

What do you think? Does your team employ these best practices for inside and outside your analysis software? If not, can you benefit from adopting them?