Back in my college days, during the spring semester of my senior year, I took a finite element analysis class. I am sad to say that after taking the class I knew little more about FEA than before. I want to believe the reason for this lack of learning was because during the class I was being trained as an operator of a specific software package. I still contend that the fact the class was held at 8AM had nothing to do with it! I had to wait until the year after when I took a graduate level FEA class (at 2PM) to learn what all those check boxes in the software package I was shown the year prior actually meant.
I was recently reading the comments on the blog post of my colleague a few weeks back and was excited to see that several people commented on the importance of calibrating discrete element models. I’d like to take those comments a step further and add that training of the user is equally as important as calibration of the model. Maybe it’s because I was tortured for years in the theory and developments of DEM, or because back when I was doing my doctoral research I didn’t have the luxury of checking boxes to run my DEM simulations (I used my own FORTRAN code), but every time I see DEM results I become skeptical. It’s how the model is calibrated that is key. If you’ve read my previous blog you know that granular materials don’t scale as easily and readily as fluids and gases. So the experience and knowledge of the operator is paramount to ensure that not only the model is calibrated but also to make sure that the calibration procedure makes sense.
So while I may cringe a little less knowing that users of DEM are recognizing the importance of calibration, I still fear that DEM users are merely operators and not granular mechanicians.