Experiments Involve Keeping Certain Respond To 2
Respond to each classmate post with at least 2-3 paragraphs…CITE ALL REFERENCES….write as if you are talking to them directly….NO PLAGIARISM
POST 1:
BA670
Week 5 Discussion Question
Design of Experiments, or DOE, is a subdivision of applied statistics that involves planning, implementing, evaluating and deducing controlled tests and experiments to assess the control factors of the value of a parameter or collection of parameters. A strategically performed experiment should provide a lot of important information about the effect on a variable due to a variety of factors. The majority of tests and experiments involve keeping certain factors constant and varying the aspects of another variable. A lot of current statistical methodologies to designed experiments are derived from the work and research of R. A. Fisher during the 20thcentury. He exhibited how taking the time to really consider the execution aspects of an experiment prior to trying it helped avoid commonly met problems in exploration. A successfully, well-performed experiment could provide information on key factors, acceptable settings, main interactions, and variations in outputs of the experiment (Bower, n.d.).
Key concepts in DOE include blocking, randomization and replication. Blocking is a concept where randomizing an aspect is impossible or is not financially feasible. Blocking allows for restriction of randomization by having all trials with one setting of the factor and then repeated with other settings. Randomization is the order in which the tests of an experiment are executed which helps eradicate effects of uncontrolled variables. Replication is repetition of the entire experiment form beginning to finish (Bower, n.d.). Because DOE is a systematic technique used to determine the relationship between issues affecting a process and the yield of said process, it is an important tool in relation to Six Sigma. More simply put, it is used to establish cause-and-effect relationships, like that illustrated in a fishbone diagram. Cause-and-effect information is required to manage process inputs in order to augment the output (Sundararajan, n.d.).
Bower, K. M. (n.d.). What Is Design of Experiments (DOE)? Retrieved July 28, 2018, from http://asq.org/learn-about-quality/data-collection…
Sundararajan, K. (n.d.). Design of Experiments – A Primer. Retrieved July 28, 2018, from https://www.isixsigma.com/tools-templates/design-o…
POST 2:
Neville Doyley
BA670
Week 5 Discussion Question
Hello all,
The DOE, not the Department of Energy but the Design of Experiments is an essential part of six sigma. When six sigma “project teams want to understand how different combinations of inputs affect the final product,” (Munk, 2015) they use this method. DOE “pulls apart the variables in a production process and evaluates the contribution that each one makes to the end product,” (Munk, 2015) allowing for a thorough and deep inspection of all parts of the mechanism. DOE does all of this in a three-step process that includes factors, levels, and responses. Depending on the project, a DOE will break down all the materials in a project and check for flaws. For example, the quality of the metal in the construction of an automobile and the length of the screws used to attach the metal components to each other during construction.
DOE is an advanced way of using applied stats to find inconsistencies in a project. It is six sigma’s deepest diving component and allows for ultimate efficiency. Running the factors, levels, and responses method will produce uncontrolled variables which can then be studied and corrected if necessary. To put it simpler terms, running a DOE will produce the “information that is needed to manage process inputs in order to optimize the output.” (Sundararajan, nd)
Neville Doyley
Munk, J. (2015, February 16). How Six Sigma Uses Design of Experiments. Retrieved July 30, 2018, from https://www.sixsigmadaily.com/six-sigma-uses-desig…
Sundararajan, K. (n.d.). Design of Experiments – A Primer. Retrieved July 30, 2018, from https://www.isixsigma.com/tools-templates/design-o…