【正文】 ation of controllable factors (independent variables) at different levels to see their effect on some response (dependent variable). 實驗 調整 可控因子 （自變量）使其處于不同的水平以檢驗對響應（因變量）的影響 y f x x x k = ( , , . . . , ) 1 2 Experimentation 實驗設計 y=f(x) from process map? 從過程圖中可以得知 y=f(x)嗎？ Six Sigma Green Belt Training Design of Experiments (DoE) Read the case material provided and answer the following as teams: 閱讀所提供的案例材料并分組回答下面問題： ? Based on a simple pmap for the curing process, what are the inputs? What is the desired output? What other questions might you ask about the process? 依據當前過程的簡單過程圖，輸入是什么？期望的輸出是什么？對這個過程你還會問哪些問題？ ? What value is the historical data – the fact that the best set point has been 155 degrees for 100 minutes for the old adhesive/hardener bination? How relevant is this information? 以往數據的產出率是多少，即老的硬化劑與粘接劑在 155度烤 100分鐘的最佳設置時的實際狀況？這個信息有多大用處 ? ? What questions might you have about the new experimental runs with the new hardener? 對新硬化劑運行的新實驗，你會有什么問題？ ? Based on what’s known at the moment, would you suspect the supplier of the new hardener as the source of the degradation in yield? Why or why not? 就目前你所知道的信息，你認為導致產出率下跌的原因是新的硬化劑嗎？為什么？ Case Study – Yield Improvement Section 1 案例分析 – 改善產出率之第一節(jié) Six Sigma Green Belt Training Design of Experiments (DoE) Read the case material provided and prepare the following as teams: 閱讀所提供的案例材料并分組回答下面問題： ? Plot the data from the first three runs. Plot yield against oven temp. Does the data support the operator’s claim? What additional questions might you have? 把首次實驗的三次運行結果畫在圖上，以烘烤溫度為橫軸畫出產出率。這些數據支持生產部門的說法嗎？你有其它問題嗎？ ? Plot the data from the next three runs. Plot yield against time. Does this data support the operator’s claim? 把第二次實驗的三次運行結果畫在圖上，以時間為橫軸畫出產出率。 Imagine there are a large number of possible factors affecting the response variable: 設想有一大堆的因子影響響應變量： ? How long would OFAT take to identify critical factors and where they should be run for best results? 靠 OFAT要多久才能找到關鍵因子和最佳效果的運行？ ? How much confidence would you have that the knowledge gained would apply in the real world? 你對 OFAT得出的結論在實踐中應用有多少信心？ OneFactorataTime (OFAT) 一次調整一個因子 Six Sigma Green Belt Training Design of Experiments (DoE) Although OFAT may simplify the analysis of results, the experimental inefficiency is significant: 雖然 OFAT可以簡化對結果的分析，但實驗效率極差： ?Unnecessary experiments may be run. The experimenter may study factors which ultimately have no impact on the output. 運行不必要的實驗。 ? OFAT experiments do not allow interactions to show themselves, since we cannot see the effect of two or more variables changing simultaneously. This means that factor “A” may not have the same impact on the output when factor “B” is high as when factor “B” low. OFAT實驗發(fā)現(xiàn)不了交互作用，因為我們看不到兩個或更多因子同時改變時的效果。 ? The amount of resources necessary to run full factorial designs can be exorbitant if the number of factors is large. Full factorials are generally