MIE360 System Modeling and Simulation
Lecture Notes
Daniel Frances © 2010
1
Lecture 8 – The Role of the Central Limit Theorem in Quantitative Validation
Suppose you completed the model to estimate the average time it takes parts to flow through the
system.
Every time you run the simulation for one week you obtain the same estimate 4.1 hrs.
You want to know if your results are in the ballpark, so you collect data for a particular day
•
you ID each individual part that entered the system in that day, say p1 to p542
•
you collect the time of arrival into the system of each part
•
you collect the time of departure from the system
•
you notice that some parts entered but have not left yet.
•
so you need to continue to monitor until all 542 items have exited
•
now you realize that you need to check if items remained in your simulation more than a week
•
you make sure you find out their exit times in subsequent days of all 542 items.
•
you then calculate for each item the time in the system t
=
exit time of item – entry time of
item
•
and finally average actual stay in the system =
Σ
t/542
Suppose t
=3.5 hrs. Not bad.
Can we stop? NO!
We have to collect more data,
Collecting More Data
For reality, we have no problem; just collect data for another 4 days, as we did the first time.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Fall '10
 D.Frances
 Central Limit Theorem, Normal Distribution, Daniel Frances

Click to edit the document details