A basic article to build awareness of the benefits of DOE and response surface methods (RSM) for process optimization.
A version of this article appeared in Chemical Engineering Progress. (chem-2.pdf 56KB) April 1998.
(Click on http://www.statease.com/pubs/ital-favform.pdf for an Italian translation 435KB.
Also see a PDF of this article as published in Ric-Mach Chimica News, http://www.statease.com/pubs/sixsigma.pdf . (sixsigma.pdf 129KB) June 2004.)
Statistical tools, especially design of experiments (DOE), provides the means for quality improvement of diammonium phosphate (DAP) and related fertilizer products. Depletion of high grade phosphate ores in Florida and elsewhere makes it increasingly difficult to meet customer specifications for nitrogen content of DAP. Urea or ammonia can be used as nitrogen supplements, but this adds cost to the final product. This paper lays out a special form of DOE, called two-level factorial design, which helped to maximize nitrogen content in DAP and make it less susceptible to impurities in lower grade phosphates.
Design of experiment (DOE) tools provide an efficient means for you to optimize your process. But, you shouldn't restrict your studies only to process factors. Adjustments in the formulation may prove to be beneficial, as well. A simple but effective strategy of experimentation involves: 1. Optimizing the formulation via mixture design; and 2. Optimizing the process with factorial design and response surface methods. This article shows you how to apply DOE methods to your formulation. A case study gives you a template for action.
Inspirational examples of DOE being applied in non-manufacturing, including sales and marketing (particularly for web-page development), but also billing, education of medical patients and many other business processes that can be quickly improved (and most effectively!) via multifactor testing methods.
The traditional approach to experimentation requires changing only one factor at a time (OFAT). However, the OFAT approach does not provide data on the interactions of factors, a likely occurrence with processes. This white paper lays out the tried-and-true "SCO" strategy of screening and characterization via two-level factorial design of experiments (DOE), followed, if needed, by response surface methods (RSM) for process optimization.
This paper discusses supplementing Lenth's method by combining his estimate of error variance from small factorial effects with the pure error variance from replicate observations or (possibly) other estimates of error variance. Presented at the Fall Technical Conference.
With the latest computer software, today's formulators can take advantage of a powerful tool: design of experiments (DOE) for mixtures. DOE methods use test arrays that produce maximum information from minimal runs. Industrial experimenters typically turn to two-level factorials as their first attempt at DOE. However, mixture design accounts for the dependence of response on proportionality of ingredients where factorial design does not. If you formulate chemicals, food or other products, consider using mixture design rather than factorials or related optimization methods. To show you how, follow along as we conduct a kitchen chemistry experiment on pound cake
In many rubber and plastics processes, powerful interactions affect final performance. These remain undiscovered via traditional one-factor-at-a-time scientific methods. Multifactor design of experiments (DOE) reveals these interactions that lead to breakthrough improvements in process efficiency and product quality. The big gains come from a very simple form of DOE called two-level factorial design. This approach to experimentation has proven to be especially helpful for control of part shrinkage as demonstrated in a case study. However, it can be applied to any measurable response in rubber and plastics production. This primer provides the essential details on two-level factorial DOE from an engineering perspective with an emphasis on the practical aspects.
Engineers at an aluminum-casting company were struggling to understand why a particular part came off the line filled with inclusions. Having conducted many one-factor-at-a-time tests to no avail, they turned to statistical software and a process called design of experiments. Optimizing based on this process let the engineers reduce the defect rate to zero.