衡量系统阐述与改善阶段关键概念.ppt

Key point This is where we are going. The Z value is the key metric for Six Sigma.. Either type of data can generate a Z value and it can be calculated from either DPMO or mean and standard deviation. Sometimes the difference is clear. Sometimes the difference is not as obvious. Statisticians never accept the null hypothesis as true. They actually fail to reject it because of insufficient evidence. Practically, if the null hypothesis is not rejected, then it is accepted as true. Conversely, if there is sufficient statistical evidence to reject the null hypothesis, then the alternate hypothesis is accepted as true. (The Minitab help system states the null hypothesis for each of its hypothesis tests.) There are two risks in a trial. First, there is the risk of convicting an innocent man. That is the risk of rejecting the null hypothesis when, in fact, it is true. The second risk is the risk of letting a guilty man go free because of insufficient evidence. That is the risk of accepting the null hypothesis when it is, in fact, false. In the courtroom example, the probability of convicting an innocent person, a, is of critical concern. To minimize the risk of convicting the innocent, overwhelming evidence is required to conclude guilt. To put it in hypothesis testing terms, to reject H0 requires a high probability (usually 95%). The terms producer and consumer risk come from the idea of accepting or rejecting material at final inspection. If the null hypothesis is true but it is rejected anyway, some material that is within specification will be rejected. This is producer risk. If some material is actually out of specification, that is, if the null hypothesis is false, but we accept it due to Type II error, then it is shipped to the customer anyway. This is consumer risk. A macro-map can be useful when reporting project status to management. A macro-map can show the scope of the project, so management can ad