Design & Factor Selection
Design Types & Categories
by Richard Scranton with Daniel Sloan
Design of Experiments (DOE) makes processes perform more efficiently, with increased quality, and decreased costs. DOE is used extensively for process development, improvement, and optimization. The information obtained through a designed experiment characterizes a process, so the process can be improved.
DOE is being used in health care to improve patient safety and to reduce health care costs. I had the opportunity to speak with Mr. Daniel Sloan about current uses of DOE as it is being implemented in the health care industry. Mr. Sloan is an independent consultant with Quality Health Systems of America, Inc.
Mr. Sloan said that DOE is selected as a quality tool because, "It is an ethical imperative to judge medical and managerial decisions using sound statistical reasoning." Using DOE, and other quality improvement tools, process factors that contribute to high quality patient outcomes can be isolated, controlled, and reduced in variability. Through this process, quality, efficiency, and cost improvements will be seen.
According to Mr. Sloan, DOE is being used for such diverse applications as investigating the causes for emergency room wait times and determining which treatments in the operating room yield the fastest patient recovery times. Typical designs involve two to three factors, and are based on qualitative models. Using the information provided by these experiments, decisions are made by statistical evidence of the outcome, not by instinct. This statistical evidence is able to assist care givers in determining which factors are most significant. With this knowledge they can streamline their process improvement goals, and efficiently enhance patient safety and improve patient outcomes.
Using contour, box/cube, half-normal probability, and interaction plots, it can be determined what factors contribute the most to the response, how the factors interact, and what factor levels provide the best response. This graphical analysis is particularly appropriate for qualitative factors typical in health care applications. By visualizing the analysis, it can be seen almost instantly the factor levels which provide the highest quality return.
The final question is how does DOE implementation enhance an existing SPC/CQI program. In the words of Daniel Sloan: "It is a geometrical and quadratic extension of advanced Statistical Process Control. DOE provides the tools to acquire a high rate of process improvement, while simultaneously producing dramatic cost reductions."
Our DOE-PC IV software product has been used by Mr. Sloan to help health care organizations improve quality and decrease costs. The user friendliness and design selection properties make it the easiest software product on the market to use for this application. Also, the graphical abilities of the software enable the user to quickly analyze experimental data, and draw statistical results from this data.
Learn more about the DOE tools for designed experiments in Six Sigma Demystified (2011, McGraw-Hill) by Paul Keller, in his online Intro. to DOE short course (only $99) or online Advanced Topics in DOE short course (only $139), or his online Black Belt certification training course ($875).