【正文】 or At A Time ? Most monly used traditional method ? Allows controlled parison of the two levels of each factor Advantages ? Easy to conduct and analyse ? Logical Disadvantages ? Not representative of “Real” conditions ? Do not know if result is significant ? Do not know the result (effect) if other factor levels were changed ? Presence of variation means misleading conclusions Run Factors Number A B C D E F G Response 1 2 3 4 5 6 7 8 1 1 1 2 2 2 2 2 1 1 1 1 2 2 2 2 1 1 1 1 1 2 2 2 1 1 1 1 1 1 2 2 1 1 1 1 1 1 1 2 1 2 3 4 5 6 7 8 1 2 2 2 2 2 2 2 1 1 2 2 2 2 2 2 EN/FAD 109 0015 169。 Ola Johansson, 1999 Ericsson Quality Management Institute DOE Objectives ? To determine those variables / predictors (time, temperature etc) that are most influential on the response ? To assess the best settings / levels of those significant variables that result in a response near a desired value maximum / minimum / nominal. ? To assess the best settings / levels of those significant variables that miminise variability in the response. ? To assess the best settings / levels of those significant variables that reduce the effect / impact of ?noise? nuisance variables EN/FAD 109 0015 169。 Ola Johansson, 1999 Ericsson Quality Management Institute Intro to Design of Experiments ? Is the process stable? You cannot accurately predict product quality (location or dispersion) without a stable process. Stability(assessed with control charts) ensures that the experimental results will provide an accurate process prediction ? What are the goals for the experiment? What factors are important? How do the factors work together to drive the process? How can you achieve optimal results from the process? These questions are actually sequential。EN/FAD 109 0015 169。 Ola Johansson, 1999 Ericsson Quality Management Institute EN/FAD 109 0015 169。 you cannot answer the last question without having the answers tp the first two. ? What is the working environment? Do you have unlimited access to the process to be able to change settings (production line Vs pilot plant access to experimentation)? How many runs can you afford in terms of time and money? Is cost a limiting issue? Will experimental results apply directly to the process or will they need verification? These are key concepts because you want to minimize the cost of obtaining information. ? What is your knowledge of the process? What do you know about important factors and how they work together (interactions)? How close to optimum are you currently running? What is the operating range of each factor? Once you define your goal, your environment and your knowledge, you can choose an adequate experimental design. Four questions to determine the type of design EN/FAD 109 0015 169。 Ola Johansson, 1999 Ericsson Quality Management Institute DOE The basics A Basic Model A body of explanatory data Measurement of Response (Potential Causes) (Effect) x1 x2 x3 y __ __ __ __ __ __ __ __ __ __ __ __ x y Independent Variable Y = a + bx parameters coefficients a b Dependent Variable A Straight Line A Statistical Model y = ?0 + ?1X1 + ?2X2 + ?3X1X2 + Error ? Linear 2 factor model ? Need to determine ? coefficients To find ? ?our estimate of real wor