DMAIC-I阶段(UT)课件.pptVIP

DOE实验设计（简述） DOE （Design Of Experiments） －General Model实验设计模型 ： Product or Process Factors (controlled variables) X1 X2 : Xn Response (measured outcome) Y Noise (uncontrolled variables) ? ~ NID(0 , ?2) ^ Y = Y + ? ^ DOE实验设计（简述） Experimental Design Terminology (术语) Response （响应） a measurable outcome of interest, e.g. yield, strength, etc Factors （因子/因素） controllable variables that are deliberately manipulated to determine their individual and joint effects on the response(s), e.g. temperature, time, etc Level （水平） a value or setting of a factor, e.g. Level 1 for time = 2 hr Level 2 for time = 3 hr Treatment （处理） a set of specified factor levels for an experimental run e.g. Treatment 1: time = 2 hr and temperature = 175°C Treatment 2: time = 3 hr and temperature = 225°C Noise （噪音） variables that affect product/process performance, whose values cannot be controlled or are not controlled for economic reasons DOE实验设计（简述） DOE – Objectives （目标） To determine which factors have an influential effect on the response To determine the appropriate settings of the influential factors for optimization of the response To determine the appropriate settings of the influential factors for minimization of the response’s variability To determine the appropriate settings of the influential factors for maximization of the response’s robustness vs the noise factors DOE – Applications （应用） Product Design：identification of key design variables；design optimization；preliminary assessment of product variability Process Design：identification of key process variables；assessment of effects of uncontrolled variables；process optimization；process capability studies Process Operation：determination of component tolerances；on-line process monitoring；on-line process optimization DOE实验设计（简述） One-Factor-At-A-Time (OFAT) Strategy 同时变量 A sequence of experimental runs in which only one factor is manipulated while the values of all other factors are held constant, followe