[R] discrete event simulation - "factory" with rework

seanpor seanpor at acm.org
Fri Jul 25 11:02:16 CEST 2008


Good morning,

I wonder could somebody help me bridge the gap?  I want to try and simulate
a "factory" type environment with an object moving between stages of
production with errors and re-work.  The intent of this is to demonstrate
the effect of error rates and re-work at various stages in the 'factory' on
the total volumes of work at each stage of the factory - in order to trigger
the change process in the 'factory' - we need to examine the end to end
process - not add in more quality control.

Going back 20 years ago I did this with GPSS, and it wouldn't be overly
difficult to write this in Java or Perl or C or... however I'd like to try
and do this in R in a vectorised manner rather than loops inside loops.  I
understand that this can be seen as a fairly simple MCMC problem and I've
created a transition matrix for my 8-state problem.  But having looked at a
number of simulation or mcmc related packages, I haven't managed to bridge
the gap in my understanding between my "need" and how to start things in R. 
The MCMC literature uses language that goes over my head.

Could somebody give me a pointer please as to where to start?

Many thanks in advance,
Sean O'Riordain

-- 
View this message in context: http://www.nabble.com/discrete-event-simulation---%22factory%22-with-rework-tp18647909p18647909.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list