So, You’ve Started Simulating—Now What?

This column is supposed to be about using simulation data management (SDM) and product lifecycle management (PLM) technologies to cover CAE data and processes. We will get there, but first we need to talk about simulation in the overall context of your design process.

Recently, I’ve spoken to a lot of CAD users who are excited about the latest shiny toys and seem to have lost interest in the oldie but goodies. Attention has shifted from CAE to additive manufacturing, IoT and virtual and augmented reality, with simulation losing its cool-kid status. That’s too bad.

If you currently over-design to avoid CAE, you probably spend more than you should on materials and shipping. If you rely on your parts suppliers to certify fit-for-purpose, you may not be as innovative you could be. People come up with all sorts of reasons to not use simulation but none are, today, defensible. CAE products are more affordable than ever—heck, structural and motion finite element analysis (FEA) is bundled into many CAD products. It’s far easier to use than it used to be and is much more intuitive. It likely runs on the hardware you already have access to and quickly produces results. With these advantages, it’s rarely going to disrupt your design process by adding significant cost or time.

Learn it, try it, then explore and understand everything you can about your designs.

Getting Results and Promoting Consistency

Once you’ve committed to a simulation program, you need to consider how you’re going to track the models and results, ensure repeatability and automate what you can. That’s where PLM and SDM come into the picture. Both began as repositories, places to put files and track access and changes. That’s tricky for a lot of simulation-related files, simply because of their size. So rather than tracking the exact files, SDMs track model references, inputs, outputs and the path from one to the other. This means that an SDM can also be used to automate simulation processes—creating process automations that ensure consistency as well as traceability.

As products get more complex and release schedules accelerate, we have less time in each design cycle to experiment with new concepts. But what if we could jump-start that next cycle with data from a prior iteration?

If we already knew that using material X would create a lighter part, but one that’s not strong enough under design criteria Y and Z, rather than running that simulation again, we could refer to prior results and move right into something new. But that presumes we can find those old runs—enter SDM.

If you can develop a consistent set of tags and other reference mechanisms to organize your simulation models and results, you can set up this data store as a valuable resource for collaboration today and for future work, too. Factor in the types of queries they might make: model parameters, material types, load conditions, solver versions, outputs—all will have relevance in upcoming projects. Your objective is to save time later by anticipating now what might be needed then.

Expanding on Previous Design Generations

Consider legacy projects, too. If, for example, you’re working on the second generation of your product, gather the simulation models and results for gen one. Start building up a repository of “tried that, worked” and “tried that, didn’t work” concepts and results. New designers can explore this treasure trove to learn how your products work. You can teach your marketing team to question the repository, as they work through priorities for the next-generation product. If they can see that a potential design feature might not be workable, they can front load their market discovery processes to find something else that could be a true differentiator. And you can use it to inform all of the simulations you do now, avoiding pitfalls unique to your designs.

SDM and PLM can be so much more than data warehouses, if you implement them with these end-use cases in mind. If you have many designers and engineers doing simulations, SDM enables you to define and deploy simulation methods across the team, automate processes as well as manage both simulation and test data. The first step is doing the simulations and storing as much about them as you can; automation and consistency are add-on benefits that you shouldn’t overlook.