evaluating FEA effectiveness

[blt01]

I am looking for metrics to help determine improvements in our FEA process. I was using "time spent per project" as a metric, but as scope and detail has increased, this metric is not that meaningful. I do 25-35 projects a year, so PPMs are probably not the most appropriate either. Does anyone on this forum have a recommended way to track their efficiency and/or performance gains in FEA? Thanks

 

[btrueblood]

The best one I could think of (given 10 seconds of thought) would be to track revisions to the original part drawings vs. amount of time spent on FEA analysis. If things are working the way they should be, you would have a negative relationship between the two.

 

[rb1957]

"surely" the true gain in using FEA is the time saved compared to hand calcs/alternative methods.

i'm willing to bet that at least 90% of your models have unique features that require learning (and re-learning) with each project.

the most meaningful metric is "did the model make the correct predictions ?" give yourself 0% for an inservice failure that should have been predicted, and 100% in the absence of inservice failures (the benefit of the doubt ?).

the most meaningful metric from a project sense is "did you meet your schedule ?"

just my 2c ...