[R] R for Engineering (Mechanical, Industrial , Civil, etc.)

[Rob Goedman]

Harsh,

Not in an educational setting, but I use R a lot in 1 branch of mechanical engineering: finite element method, as in Przemieniecki's recent book.

R prepares the input for multiple runs of a fortran based FEM routine and searches for an optimal solution (position of a bottom hole assembly in a deviated bore hole based on a set of constraints). The fortran routine in turn generates R source code/data which is then read in by R and used to formulate the input for the next run (set of FE iterations) and, once a converged/compliant solution is obtained, for graphing the results.

I started off with a version just in R but that turned out to be too slow and redid the FEM part in fortran (It should be said it was my 1st serious attempt to use S4 and I'm sure I could have improved the R-only performance, in a future version I will likely return an S4 class definition/object).

The ability to include fortran, C, etc., R's graphical capabilities, including 3D, and R's capabilities/libraries (e.g. different interpolation models) to generate the input files based on survey data and revising the input during iterations were key for me.

Regards,
Rob

On Apr 25, 2010, at 2:53 PM, Harsh wrote:

> Hi useRs,
> In trying to take R to engineering undergraduate students, I have been
> looking for context that would make R more accessible to the said
> audience. Though R is primarily a statistical tool, I would want to
> demonstrate the use of R for certain engineering courses (Design of
> Machine Elements - gear design, ball bearings, etc.) which would
> generate interest in it and provide students a way to extend its use
> to more "statistically" oriented courses like Quality Control (control
> charts, t-tests, drawing normal probability plots, etc.).
> 
> For example: Engineering Design has a topic that requires the solving
> of a set of equations to arrive at the appropriate parameters of a
> feasible gear design (diameter, number of teeth, material
> specification etc.). This would require the set of equations and other
> information to be available in a package, with added functionality to
> plot machine elements, reliability curves for various materials (steel
> of different carbon compositions, and such) and other
> visualisations/computations that are required for such studies.
> 
> If any R user has employed R to teach engineering courses (which do
> not require much statistics), I would highly appreciate your feedback
> and insights gained from such an undertaking.
> I am not trying to dilute R's primary focus in being a statistical
> tool, but I would like to make R available to an audience who do not
> deal with a lot of statistics.
> 
> Of course, there are other tools for engineering drawing, circuitry
> design and such others, but maybe there is a niche area (somewhere in
> between core engineering and statistics) which is yet untapped, and R
> "might" be of help there.
> 
> Thank you.
> 
> Regards
> Harsh Singhal
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.