DOE (Design of Experiments) in Development Chemistry:  Potential Obstacles

Abstract

There are often issues associated with challenging accepted working practices through the introduction of statistical tools. This paper outlines common objections to the use of DOE (design of experiments) and our standard responses to overcome these obstacles.

Hydrodynamic effects on mixing and competitive reactions in laboratory reactors

“…This approach was useful to study the main effect of the variables on the outcome of a mixing sensitive reaction system. It also allowed for the study of how these variables interact with each other, something that is not easily accomplished using the traditional one-variable-at-a-time approach (Lendrem et al, 2001). The experimental results were interpreted in the light of computational fluid dynamics (CFD) simulations, which provided information on the velocity distribution and energy dissipation rate in the reactors.…”

Mesh Morphing and Response Surface Analysis: Quantifying Sensitivity of Vertebral Mechanical Behavior

“…There is no a priori reason to expect that a second order polynomial shall provide an adequate representation of the system, however, in practice second order polynomials often work. 6,27,29,37,52,54 The regressed function found through an RSM fit is not a functional relationship, but rather an approximation within defined bounds. The analysis provides a means to evaluate quality of fit and determined that the choice of the function was adequate for characterizing the caudal vertebral biomechanics.…”

“…4,16,32 A response surface method (RSM) is a more efficient method to determine factor influences, independently and in interaction, as has been demonstrated in its application in many fields. 4,6,8,17,27,29,31,37,52,54 A further advantage of an RSM approach is that it produces metamodels, which are analytic phenomenological descriptions of the response as a function of the factors. We propose that it is possible to combine the benefits of generic and specimen-specific modeling through the use of morphing techniques to alter the geometry of a FE model in a controlled and systematic way suitable for parametric study.…”

Optimization of a Sonochemical Process Using a Novel Reactor and Taguchi Statistical Experimental Design Methodology

“…The technique of defining and investigating all possible conditions in an experiment involving multiple factors is known as the design of experiment (DOE). [35][36][37][38][39][40][41][42][43][44] The classical method (full factorial design) used in statistical design of experiments requires a large number of experiments to be carried out when the number of process parameters increases. For a full factorial design, the number of possible designs or experiments N is N ) L m , where L ) number of levels for each factor and m ) number of factors.…”

“…For a full factorial design, the number of possible designs or experiments N is N ) L m , where L ) number of levels for each factor and m ) number of factors. [35][36][37] To solve this problem, the Taguchi method uses a special design of orthogonal arrays to study the entire parameter space and determine the optimal design with only a small number of experiments. The advantages of this method are that it has standardized factorial design of experiments and saves a lot of time and expense while investigating the contribution of the various factors.…”