Histogram Interpretations
Six Sigma – iSixSigma › Forums › General Forums › General › Histogram Interpretations
 This topic has 8 replies, 5 voices, and was last updated 11 years, 8 months ago by Akbar.

AuthorPosts

August 24, 2010 at 7:58 am #53556
AkbarParticipant@naliakba Include @naliakba in your post and this person will
be notified via email.Hi there,
I want to use Histogram to arrive on standard time of a given process. This will be easy if the data is normally distributed (symmetric). I have some cases where it is flat, bimodal, skewed etc.Can someone help me in getting guideline in this regard. Pl. note that this standard is going to be base for me to arrive upon headcount required to perform a particular activity.
0August 25, 2010 at 1:39 am #190665I don’t know what you are trying to do with the histogram, but why not simply sum up all work content (ie all the work elements or tasks required to complete a unit of work for the given process) and divide through by the projected takt time. That will give you a good idea of required staffing levels. I would look to the LEAN tool kit here, as you might find Value Flow Analysis and subsequent standardization helpful in terms of labor requirements.
0August 25, 2010 at 1:42 am #190666Actually, a planned cycle time instead of Takt would be preferred in the aforementioned calculation. Planned Cycle Time = Takt * Usage Rate (eg .85.95 of takt time).
0August 25, 2010 at 10:03 am #190667Send me the Data in Excel sheet, I will analyse the Histogram and suggest you the answer
0August 26, 2010 at 3:51 pm #190671
AkbarParticipant@naliakba Include @naliakba in your post and this person will
be notified via email.I will try with TAKT and Planned Cycle time approach. In the meantime, attached the sample data for 4 independent processes. [file name=Histogram_Interpretations.xls size=32768]https://www.isixsigma.com/images/fbfiles/files/Histogram_Interpretations.xls[/file]
0August 27, 2010 at 10:45 am #190672To be honest, I don’t think there is any magic answer. I would go out to the factory floor and do time studies so I could understand the data. Maybe it is bimodal because the second shift has fewer distractions (No managers around).
0September 1, 2010 at 10:32 am #190687
Roger EllisParticipant@Mellophone Include @Mellophone in your post and this person will
be notified via email.naliakba wrote:
Hi there,
I want to use Histogram to arrive on standard time of a given process. This will be easy if the data is normally distributed (symmetric). I have some cases where it is flat, bimodal, skewed etc.Can someone help me in getting guideline in this regard. Pl. note that this standard is going to be base for me to arrive upon headcount required to perform a particular activity.
You need to consider the variation that is present in your data. For example, assume a distribution with normal data. If you use the average time as your standard time, 50% of the workers will be able to meet or exceed the standard, and 50% will be in the opposite circumstance – they will not be able to get the work done in the allotted time.
You must decide what percentage of the population that you wish to be able to consistently perform the job. Then, use your knowledge of the mean and the standard deviation of the population to select a standard time that is longer than the average such that most of the population will be able to perform the job within the allotted time.
Regards, Roger Ellis
Six Sigma Master Black Belt0September 6, 2010 at 4:17 pm #190703To arrive at Standard Time for a process, the best tool would be Standard Work Lean Tool.
Normal Distribution is symmetric about mean. In your case mean would be Standard Time. The Std. Dev (sigma) will give you the observed variation which you have seen while tabulating data.
Skewed or Bimodal does not change the above scenario because in Normal Distribution Mean, Mode & Median are all same while in Skewed /Bimodal, Peak of the Graph will not happen at Mean but at Median or Mode.So don’t bother. Use Lean method to get Standard Time.
This will also help you in minimizing NVA activities. Normal Distribution will not help in reducing NVAIf you still want to use data distribution, then take Descriptive Statistics of the data collected which will give you mean, median and Std.Dev.
Use Mean or Median + Std Dev for Upper Limit and – Std Dev for Lower Limit.Regards,
Anupam0September 8, 2010 at 8:09 am #190706
AkbarParticipant@naliakba Include @naliakba in your post and this person will
be notified via email.Thanks to all who contributed in this discussion forum.
0 
AuthorPosts
You must be logged in to reply to this topic.