Sixsixma basic analysis example

CASE STUDY I
PUDDY COMPUTERS
IOE 461
QUALITY ENGINEERING PRINCIPLES AND ANALYSIS
UNIVERSITY OF MICHIGAN
PROF: PAT HAMMET
{draw:frame} {draw:frame}
(Double Click to see content)Alan Aldrete: alan.aldrete@gm.com
Carmen Rodriguez: carmen.rodriguezleon@gm.com
Submit Date: 1OCT09
DEFINE Problem Statement:
Data from PC shows that many of their custom ordersare above 28 business days resulting in loss of competitiveness in the market and extra cost due to extra inventory, facility space storage, internal tracking, and shipping cost. This has been achronic problem and it is costing over $44K in the last 3 months (estimated over $160K/year)
Goal Statement:
The goal is to have all custom orders filling time to be less or equal than 28 businessdays.
Project scope:
Focus on reducing the custom order filling time between stages: Order placement through shipping date.
MEASURE Historical data from the last 3 months shows thatmany custom orders are above 28 days, indicating that this is a chronic problem see Chart 1.
The current state is described on the Table 1:
It is clear that Puddy Computer have issues with thecustom order filling time, since the Yield is away from the 99.94% and the DPM is much more than 3.4 for a world class company.
As additional information, it can be seen on the Histogram, Chart 2,that there is a great variability in the process with many of the custom orders above our USL= 28 days.
It can be seen in the Histogram (Chart 2), as well, that the data is not normally distributedindicating that the actual measuring system is not so accurate (P-value < 0.05 and Skewness = 0.76- skewed to the right)
Sample data shows a normally distributed behavior (P-value=0.1) indicating abetter measurement system, what give us more confidence in the results obtained in this study. It is shown in Chart 3A and Chart 3B the comparison between both set of data.
Following Box Plot...