05-04-2014 10:19 AM
Is it possible to define a distribution in a table for the processing time in a singleproc. I am currently using a table to to define average processing time for several different parts but I would like to give each part its own lognorm or normal distibution. If a table is not the right way to do it what would be?
05-04-2014 11:52 AM - edited 05-04-2014 11:53 AM
Yes it is possible. Normal distribution needs some parameters: Stream, MU, Sigma. Lower bound and Upper bound. You have to define this values for each part in a table. To set the value of ProcTime you can use the option Formula. You can access the MU with @. If you set the MU names a row-index in the table the formula could look like this:
z_normal(procTimes[1,@.name],procTimes[2,@.name],procTimes[3,@.name],procTimes[4,@.name],procTimes[5,@.name])
Look here for an example (Version 11):
Steffen Bangsow freelance simulation specialist web: www.bangsow.eu mail: steffen@bangsow.net | ![]() |
05-04-2014 12:00 PM
Oh, thats easy. My approch was all wrong thank you.
03-29-2017 12:01 PM
How would we expand the example you provided to have multiple singleProcs with different mu and sigma?
Say i have Op1, Op2, Op3.
03-29-2017 12:14 PM
For instance by selecting "normal" and adding individual parameters (mu/sigma) for each station
03-29-2017 12:22 PM
The table containing the mu, sigma, upper and lower bounds will have rows for each Operation with specific mu and sigma. I have attached a picture of the table.