I'd argue that software development is much more like other engineering disciplines 
than, say, fiction writing or athletics.
Mechanical design is an area still very much like software design.  There are are high 
level designers, sometimes called "engineers" and
lower level designers, the people who used to be draftsmen, but now use CAD 
workstations.  This is because there is still far too much
low level detail work to make it cost effective to employ the most highly 
trained/skilled people for all tasks.

Low level designers can wreck havoc in mechanical design.  If the hole for the 
mounting bracket is modeled 1 cm to the left of where it ought to be, you may end up 
having a $10,000 paper weight instead of a die.  Worse still, if you shipped the model 
off to a subcontractor without checking what came back from the pilot run, you may own 
100,000 paper weights.  This happens a lot more than most people think; "material 
recovery" is a euphemism for what to do with all the paper weights.  On the other 
hand, designers often have more experience with actual manufacturing processes and 
environments than engineers, and they may catch types of errors that the mechanical 
engineers miss.

In mechanical design processes, there is typically a review step, but often it is 
primarily a standards compliance review step - has the designer put stuff on the right 
layers on the drawing, etc. Less often, there is some subtantive content checking.

The time lapse between committing a design error and the discovery of the error can be 
quite long. A lot of recalls in the automotive world are probably due to design 
errors, and it can take two or three years to discover the problem.  Design errors in 
bridges or buildings can take even longer to appear.

In the mechanical world, the use of simulation is playing an increasing role in 
avoiding design errors. I note that simulation is really just a variety of testing 
with the same problems in determining what/how to 
test.  The mechnical testers have the advantage of much more powerful computer 
hardware and much flashier graphics.

Also in the mechanical world, failure analysis is a recognized subdiscipline - read 
Petrovski's books.   This may have evolved because
mechanical system failure can arise either in the design of the product or in the 
manufacturing process.  Failure analysis engineering may
have evolved to settle the disputes between product engineering and manufacturing 
engineering as to who was to blame.

Ruven Brooks
 
----------------------------------------------------------------------
PPIG Discuss List ([EMAIL PROTECTED])
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/

Reply via email to