Typical business problem
Six Sigma – iSixSigma › Forums › Old Forums › General › Typical business problem
 This topic has 3 replies, 3 voices, and was last updated 16 years, 4 months ago by Stephan.

AuthorPosts

July 6, 2005 at 12:01 pm #39913
Pankaj GuptaParticipant@PankajGupta Include @PankajGupta in your post and this person will
be notified via email.I want some feedback on a typically problem which i face while working on the data related to banking industry.
The mean time to process a case has been reduced from 1.11 days to 0.15 days
The standard deviation in the data has been reduced from 2.5 to 0.50
The upper specification limit given by the champion of the project team is 1 day and the lower specification limit has been assumed as zero by the project team.
Both the mean time taken and process varition has reduced significantly in the project. However, when i put the values of mean and standrad deviation in the sigma template the sigma level shows the following
Mean Standard Deviation LSL USL Sigma
1.11 2.5 0 1 1.5
0.15 0.5 0 1 1.8
However, if i don’t put the lower specification value as Zero the Sigma value shoots up to 3.2.
Since both the mean and the variation has been reduced (standrad deviation reduced by 500% I feel that while calculating the sigma I should not put the LSL as zero in the sigma calculator since with the LSL as zero there is no significant improvement in the sigma level despite significant reduction in process variation.
Please suggest.
0July 6, 2005 at 12:37 pm #122587You can put the 0 days LSL as boundary limit due to the fact that you will not be able to go lower then zero.
0July 7, 2005 at 4:21 am #122674
ChaturvediParticipant@Pankaj Include @Pankaj in your post and this person will
be notified via email.Yes i can put 0 as LSL . However, how do i explain that there is only a marginal increase in sigma level wherease the variation in the process has decreased by 500%. The team will be highly demotivated since they have made significant chnages in the process which has helped them to reduce the mean and the standrad variation.
0July 12, 2005 at 12:12 pm #122923I think that the calculation without setting LSL as boundary limits shows you potential out of specification for the LSL, which in fact is not possible. Setting the LSL as boundary has the same effect like eliminating the LSL. The capability calculation is then only towards the upper specification.
0 
AuthorPosts
The forum ‘General’ is closed to new topics and replies.