This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: C p = (USLS)/6 = (100.599.5)/(6x0.414) = 0.40; Not capable because C p < 1. C pk = Min{(US)/3 , (LS)/3 } = Min{(100.599.765)/(3x0.414), (99.76599.5)/(3x0.414)} = 0.725; Not capable because C pk < 1. Since C p < 1, the variability (std. dev.) must be reduced. Further, the mean must be increased, since C pk < 1. 4a. A pchart should be used because control limits are required for the proportion of wrong reports. Also, the sample size and # defective are given. b. = 20/2000 = 0.01, n = 200, z = 3 = = 0.007 UCL = + z = 0.01 + (3x0.007) = 0.031 LCL =  z = 0.01 (3x0.007) = 0.011 0 c. Sample p 1 1/200 = 0.005 2 3/200 = 0.015 3 3/200 = 0.015 4 5/200 = 0.025 5 2/200 = 0.01 All 5 samples (100%) lie between control limits. We need at least 99.7% with 3sigma limits, so the process is in control....
View Full
Document
 Fall '07
 Eng

Click to edit the document details