过程能力分析.ppt

The Z-value is a means of comparing different processes and different CTQs. The higher the Z-value, the better the process. The Z-value can be used to determine total DPM for a process. As a business, we use Z-values to talk about processes instead of DPMO much the same way we talk about order accuracy instead of order mistakes, or on-time deliveries instead of late shipments, . . . . 47 For attributes data life is much simpler. The Z score or “sigma” score relates directly to the proportion defective. Remember we don’t have a mean or std. deviation like we do with variables data. For attributes data the Z score does not relate to the # of std. deviations between the mean and specification limit. It is just a way of converting DPMO to a Z score so that we can use the same common business language. Think back to the variables data exampleswhat are we using the mean and std. deviation for anyway? To determine the total proportion defective. First, we calculate the upper and lower Z score. Then we look the proportion defective up on the table for upper and lower. We add the stuff that is out of spec on the bottom to the stuff that is out of spec on the top. Now we are at the same point that we start at with attributes data. We now the proportion defective and can use this to calculate DPMO and/or a Z score. Back to attributes data. You know from your data what proportion is defective. To determine. the sigma score you look up the proportion defective in the table and locate the corresponding z score. For attributes data life is much simpler. The Z score or “sigma” score relates directly to the proportion defective. Remember we don’t have a mean or std. deviation like we do with variables data. For attributes data the Z score does not relate to the # of std. deviations between the mean and specification limit. It is just a way of converting DPMO to a Z score so that we can use the same common business language. Think back to the variables data