The general population may not usually give a second thought to the amount of research, experimentation and implementation that goes into a purchased product, but if you work in product development, quality, manufacturing, labs or R&D, you don’t have that luxury. Those in the manufacturing industry need to concern themselves with all factors that could potentially lead to product and manufacturing litigation or warranty issues.
An effective and efficient experimentation process is the key to building good products and avoiding claims over their quality, and that starts with using statistical tools like design of experiments (DOE) and fractional factorials in your manufacturing process, which allows you to solve problems more quickly with better processes.
Avoid the One-Factor-at-a-Time Method
Experimentation is frequently performed using trial and error approaches, which are extremely inefficient and rarely lead to optimal solutions. Furthermore, when the effect of multiple variables on an outcome (the “response”) needs to be understood, “one-factor-at-a-time” (OFAT) trials are often performed. Not only is this approach inefficient; it also inhibits the understanding and modeling of how multiple variables interact to jointly affect a response.
“Statistically-based design of experiments (DOE) provides a methodology for optimally developing process understanding via experimentation. DOE, which global quality association American Society for Quality (ASQ) defines as a “branch of applied statistics [that] deals with planning, conducting, analyzing and interpreting controlled tests to evaluate the factors that control the value of a parameter or group of parameters,” has numerous applications, including fast and efficient problem solving (root cause determination), shortening of R&D efforts, optimizing product designs and manufacturing processes, developing product or process specifications, and improving quality and/or reliability.
Fractional Factorials to the Rescue
A structured approach to the experimentation process starts with an initial screening phase where you narrow your variables and identify important factors. It’s during the initial screening process that fractional factorial design experiments take place. Working on a subset of the full factorial design helps you focus your efforts on the critical part of the problem and reduces the amount of time, energy and resources you need to expend. Fractional factorial design experiments are simply invaluable when a large number of factors must be investigated.
The key concepts in creating a designed experiment, according to ASQ, include:
- Blocking: When randomizing a factor is impossible or too costly, blocking lets you restrict randomization by carrying out all of the trials with one setting of the factor and then all the trials with the other setting.
- Randomization: Refers to the order in which the trials of an experiment are performed. A randomized sequence helps eliminate effects of unknown or uncontrolled variables.
- Replication: Repetition of a complete experimental treatment, including the setup.
“DOE lets you assess the main effects of a process as well as the interaction effects (the effect of factor A, for example, may be much larger when factor B is set at a specific level, leading to an interaction),” according to a DOE trainer writing for statistical software company Minitab. “In science and in business, we need to perform experiments to identify the factors that have a significant effect. The objective of DOE is to reduce experimental costs—the number of tests—as much as possible while studying as many factors as possible to identify the important ones.”
Learn Fractional Factorial Strategies
Join statistician and quality expert Steven Wachs when he explains the use of fractional factorial experiments in an audio conference for AudioSolutionz, “Fractional Factorial Experiments for Screening Studies.” During this session he reviews the key concepts behind design of experiments, along with the strategy to utilize sequential experiments to most efficiently understand and model a process. Steve emphasizes fractional factorial studies as useful tools in the screening phase of experimentation, and he provides a valuable case study involving optimizing a manufacturing process with multiple responses.
