CAE Without CAD is Lame, CAD Without CAE is Blind

As well as dreaming up innovative ideas, small engineering companies and consultancies need access to cost-effective tools that let them explore, optimize, and validate their widgets. There are plenty of Computer-Aided Design (CAD) tools available, but CAD is not enough.

Albert EinsteinAlbert Einstein

Engineers have to optimize and validate a widget, not just draw it. Optimization means assessing (preferably with quantitative measures) an array of alternate concepts in order to select the best widget. Validation means assessing whether a widget can meet its functional requirements. Previously, costly physical experiments prevented all but the most well-funded companies from optimizing and validating their designs. But now, with Computer-Aided Engineering (CAE) analysis tools, such as Computational Fluid Dynamics (CFD) , the smallest companies and consultancies can join the party.

Needless to say that CAE is useless without the geometry provided by CAD.

To misquote Einstein, "CAE without CAD is lame, CAD without CAE is blind."


couldn't agree more...feel like a broken record

Hi Richard.. I always knew that Einstein was a CFD dude at heart.

It is interesting to talk to companies that agree with your sentiment. They get the idea that it should be an evenly flowing process from concept to manufacturing, the entire system integrated very nicely where material properties are assigned, meshing intelligence can be captured and reporting of critical data is a breeze. But, due to the nature of things, many companies can't always reach Nirvana, but strive to get close.

Then.... there are those that simply refuse to consider it and "we don't do it that way" mentality cripples their perception of efficiency.

As most know, I am a proponent of upfront CAE. Many confuse this to be "design level" simulation. That is only a part of it. Upfront CAE is really process-driven simulation- incorporating CAD, simulation processes, reporting of data etc...

Companies have to get out of the idea of "autopsy" simulation where its done after the fact and used to find the problem. Finding the problem, shouldn't be the driver of simulation. Finding the optimal, innovative design should be the driver. Simulation is a design tool, I think we forget that sometimes.

This involves the entire team. Vision is often the piece of the pie that is missing. We need engineering leaders, managers, designers to speak up and educate one another. Surely, not implying simulation is the end all be all. It's a tool, like anything else. It just needs to be used fully.


I once heard a talk by one of the founders of MSC in which he said that CAD has turned into geometric design when what we really need to be practicing is functional design. This summarizes your post nicely - we design things to be functional and their shape is simply the output - a necessary output, but an output nonetheless.

Importance of Best Practices and Lessons Learned

I agree that it is wonderful to have the process from CAD, to gridding, to CFD bundled up. However, IMO, access to a set of best practices, lessons learned, and verification and validation runs is the most important thing. This is one area I feel the OpenFOAM community has a long way to go. Regardless of how easy the process is, it still takes an expert to run CFD, an individual who knows how to create a good grid, how to set the parameters correctly for a given case, and what to look for in the solution to identify problems. It can be very easy to get a bad answer with CFD, especially with Navier Stokes runs. And there are plenty of companies who have been burned by this.

Roving Robot Report

A roving robot report on the relationship between CAD and CAE analysis tools.