statistical question
Six Sigma – iSixSigma › Forums › Old Forums › General › statistical question
 This topic has 5 replies, 4 voices, and was last updated 14 years, 9 months ago by TonyBo.

AuthorPosts

February 22, 2008 at 7:57 pm #49419
bjkeith75Participant@bjkeith75 Include @bjkeith75 in your post and this person will
be notified via email.Hello all,
I’m looking for a calculation to help me with the following –
We have a process with about 150 steps in it. The project lead, instead of collecting real data, is doing “best guesses” for cycle times and decision diamonds. I want to be able to show the real accuracy of the final number. I would expect the overall accuracy to be very low, even at a 95% accuracy on each individual step. Thanks in advance for the help.0February 22, 2008 at 8:28 pm #168906
DatalyzeParticipant@Datalyze Include @Datalyze in your post and this person will
be notified via email.First, you have to collect the actual data in a reliable manner. In the end, how you collect the data is up to you, but it directly affect the accuracy of your final calculation. The data collection process may be very time consuming since you have many steps in your process [maybe you should narrow this time by targeting key areas in your process]. Once you have collected the data, only then can you decide, (or anyone else for that matter) what calculation you should use. A little more insight into your data or process may shed more light on what you are seeking and maybe I can be of more help. Remember, a “best guess” is never a substitute for actual data.
–Datalyze0February 24, 2008 at 4:50 pm #168937
Eric MaassParticipant@poetengineer Include @poetengineer in your post and this person will
be notified via email.Hi BJKeith,
I have an idea / suggestion for your consideration:
Can you ask the project lead to obtain three estimates (most likely, practical worst case, practical best case) for each of the steps rather than just one estimate? Then, you could build this into a Monte Carlo Simulation model for the total cycle time, and compare that to the distribution of the actual total cycle time.
Best regards,Eric0February 25, 2008 at 11:44 pm #168993
bjkeith75Participant@bjkeith75 Include @bjkeith75 in your post and this person will
be notified via email.Thanks for the response Eric. While this wasn’t what I was looking for, you gave me some new insight into the situation that will be of help in the future.
I still wouldn’t mind if anyone else could answer the original question (maybe I didn’t word it well). I have 150 boxes. Each box is 95% accurate. What’s my margin of error on the average cycle time?0February 26, 2008 at 1:02 am #168996bjkeith75..
You dont have enough information….you need:
N= number of boxes (150)
S= Standard deviation (of the cycle times you took)…
and then run it in the equation: Squareroot((1.96*S)*((1.96*S)/N))
that will give you your precision.
Hope this answers your question.0February 26, 2008 at 1:15 am #168997Actually I read your initial post wrong…..I think you are way off what you are looking for. 150 steps, and you want cycle times for each step. Then take the equation I gave you, in an earlier post…and you have to take a sample of at least 10 timings per step…take that standard deviation….and the number of boxes you want/ intend on timing (your sample)…this will give you an idea of your precision. In one of your posts…you equate boxes to steps…not sure you meant to do that…unless, each step produces a box. Anyways..hope this helps…
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.