某公司的六西格码培训教材.ppt

1. 0.23 1 - 0.023 = .977 2. 0.0582 1 - 0.0582 = 0.9418 3. 6.41 - 6.34 / 0.03 = 2.33 Area to right = 0.0099 Area to left = 0.9901 For a 2 sided specification I have to do a little more work because I have the chance of having defects on both sides. Therefore I need to first calculate an upper and lower Z value. Use the table to find the proportion defective for both the upper and lower. Add the two together to find out the total proportion defective. Look up the total proportion defective on the Z table to determine the sigma score and/or I multiply the proportion defective by 1,000,000 to get DPMO. This is mathematically how we calculate Cpk and Ppk. Notice the only difference between the two formulas is the estimate for standard deviation. Also notice that to calculate Cpk and Ppk, you must calculate the upper and the lower and take the smaller of the two values. The capability indicators are : Cp Cpk or Pp Ppk. Let’s explore these indicators in more detail over the next several pages. Some really rough guidelines to go by. They come from the AIAG (Ford, GM, etc.) These will change as we get better processes and get closer to 6 sigma. A Six Sigma capable process would have a Cp value of 2.0 and a Cpk value of 1.5. Pp and Ppk can never be higher than Cp and Cpk mathematically. The guidelines for Pp and Ppk are higher because there can be situations where the tolerance is so wide that a process can be in spec and “out of control”. Under these conditions customers are sometimes willing to accept “parts” from “out of control” process so long as they are well within the tolerance...more within the tolerance then if the process was “in control”. Cp = USL - LSL / 6 s = (50-46)/ 6(1.11) = 0.60 = Cp Cpl = X - LSL / 3 s = (48.68 - 46) / 3 (1.11) = 0.80 = Cpl Cpu = USL - X / 3 s = (50 - 48.68) / 3 (1.11) = 0.40 = Cpu = Cpk Cpk = Min {Cpl, Cpu) GE Data p = 0.00, non normal data; Cp = 0.;28, Cpk = 0.26; DPMO (LT) = 45