Analyze Screening Design (Definitive & Plackett-Burman): Illustration with Practical Example

hello friends in the last video we had seen to create screening design definitive as well as placket Behrman with a practical example in this video we are going to learn the analyze of the definitive screening design with the help of the same example seen in the last video so let's begin analyze definitive screening design use analyze definitive screening design to analyze a designed experiment and identify the most important model terms as well as their interactions that can impact the response you can analyze your design after you create or define a screening design and at at least one response column in your worksheet before going for data collection we must know the data considerations of it let's see in detail data considerations for analyze definitive screening design to ensure that your results are valid consider the following guidelines when you collect data perform the analysis and interpret your results 1 the data must include at least 2 factors which can be either continuous or categorical a designed experiment in Minitab must have at least two factors that are either continuous or categorical if you have only one categorical factor and no continuous predictors use one-way ANOVA if you have one continuous factor use fitted line plot to the response variable should be continuous if the response variable is categorical your model is less likely to meet the assumptions of the analysis to accurately describe your data or to make useful predictions if your response variables have two categories use fit binary logistic regression if your response variable counts occurrences such as the number of defects use fit plus own model 3 ensure that the measurement system produces reliable response data if the variability in your measurement system is too high then your experiment may lack the power to find important effects for each observation should be independent of all other observations if your observations are your results might not be valid consider the following points to determine whether your observations are independent if an observation provides no information about the value of another observation the observations are independent and if an observation provides information about another observation the observations are dependent five the experimental run should be randomized randomization reduces the impact of uncontrolled conditions on the experiment results randomization also lets you to estimate the inherent variation in materials and conditions so that you can make valid statistical inferences based on the data from your experiment in some situations randomization may lead to undesirable run order for instance factor level changes can be difficult expensive or take a long time to produce a stable process under these conditions you may want to randomize with a split plot design to minimize the level changes 6 collect data using best practices to ensure that your results are valid consider the following guidelines make certain that the data represent the population of interest collect enough data to provide the necessary precision record the data in the order it was collected and 7 the model should provide a good fit to the data if the model does not fit the data the results can be misleading in the output use the residual plots the diagnostic statistics for unusual observations and the model summary statistics to determine how well the model fits the data example to analyze definitive screening design let's continue the same example and use the same run order to collect the data during data collection we need to follow all data considerations we had seen now a group of Engineers wants to investigate the effects of seven factors on the power output of an ultrasonic cleaner after the engineers collect the data they enter the response data in an empty column in the worksheet and use the option of analyze the design to analyze definitive design please follow the steps in minutes have 18 and 19 1 select stat design of experiments screening analyze screening design 2 in responses enter power output 3 click terms 4 and include the following terms select linear and then click OK 5 click graphs 6 select an option to display all terms 7 under residual plots select 4 + 1 8 click OK in each dialog box to get the results for analyzing screening design if you are using older versions of minute tab like 17 16 or lower then there is little different procedure due to the non availability of separate option for screening design in that case we have to use an option of analyze factorial design interpretation of the results in the Pareto chart terms that are in the model have blue bars and the terms that are not in the model have grey bars the engineer sees that the largest main effects are for train a and quiet D from the screening experiment the engineer concludes that these two factors deserve the most consideration for further analysis in these results the main effects for train and quiet are statistically significant at the point 0 5 level having P values of point zero zero four and point zero one seven respectively we can conclude that changes in these variables are associated with changes in the response variable we need to consider all terms because sometimes model terms are not significant but their interactions with other model terms can be significant in these results the model explains 72 point 56 percent of the variation for these data the R square value indicates that the model provides a good fit to the data if additional models are fit with different terms then use the adjusted r-square values and the predicted hour square values are used to compare how well the models fit the data looking at the value of r square predicted we cannot use the conclusion from this analysis to predict the response variable outside of the considered levels use the residuals versus fits plot to verify the assumption that the residuals are randomly distributed and have constant variance ideally the points should fall randomly on both sides of 0 with no recognizable patterns and the points in our example there is no such pattern indicates that the residuals are randomly distributed and of constant variance this is all about to analyze definitive screening design with the help of a practical example we will see the next important topic related to the design of experiments with a practical example in the next video for references I have taken some part of this detailed content from minute AB now to end please like this video if you have found it useful and your valuable comments and share this video with your friend and colleagues to improve and refresh their knowledge if you want to get updates of such videos from our Channel please do not forget to subscribe to it click on the bell icon and select to get all notifications and finally thank you for watching [Music]

Channel: LEARN & APPLY : Lean and Six Sigma

Views: 4,928

Rating: 5 out of 5

Keywords: DOE, Design of Experiments, D.O.E., design of experiments minitab example, interpretation of doe results, six sigma, DOE Minitab, Create Screening Design, definitive screening design minitab, plackett burman design minitab, plackett burman design example, Screening design example, Comparison of Definitive and Plackett Burman design, Advance tools in Six Sigma, Analyze Screening Design, Analyze Definitive Screening Design, Analyze Design of experiment

Id: grliM2dAET8

Length: 7min 55sec (475 seconds)

Published: Sat Jun 27 2020