On Mon, 26 Feb 2018 11:44:07 +0000, Chris J Dixon
wrote:

Jim Lesurf wrote:

I was wondering how many professional engineers or scientists do so. e.g.
use tools like MathCad, etc, whilst having no real idea how they work or of
the commands they gave are the correct/relevant ones and can be expected to
give an appropriate 'answer'. More than once I've found (professional)
people engaging in GIGO.

Quite! It is clear that if you work to the appropriate Quality
Standards that you should be able to demonstrate that you have
verified that the calculations you perform give correct results,
and that input conditions and outputs are recorded auditably.

I recall one situation where a train performance program was
being used to work out the rating for a works locomotive, and it
was realised that altering the gradient of the tunnel in which it
was to be used had no impact on the output.

The relative ease of producing a spreadsheet can lead to
automatic belief in its results, and if the parameters keep being
altered, how easy is it to back track later to find where the
numbers came from?

Chris

I make a lot of use of Mathcad. But part of the process is exercising
it with corner cases and variable sweeps to make sure that everything
it does makes sense in terms of the model.

Where possible I will always use units as they provide inbuilt
dimensional analysis. If an answer that is supposed to be in Hz
insists that it is kg/foot, I know I need to do a bit of checking.

d

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus