Optimum mixture designs for product formulation-A designs for six sigma approach

Industrial products consist of two or more components in defined proportions. Due to the competitive market conditions and quality improvement necessities, industries have to rationalize their products and production processes. There are lots of accumulated process and product improvement methodologies. Among them, Lean Six Sigma (LSS) methodologies have gained a grand interest because of joining qualitative and statistical tools together.Industries can decrease their process variances and control all of their processes via LSS methodologies. The most important tool of LSS is the design of experiments (DOE). This tool can be used for the improvement part of LSS and can also be used for the product design/development cycles.The quality of an industrial product depends on the raw material proportions and the process variable levels, both of which need to be taken into account in designing a product. A specialized form of DOE?s which can be used for designing products are named as mixture designs. By adding process variables to these kinds of designs, a new version called as mixture-process designs has been developed.Universal colorants are used in paint and related industries. In this study, after explaining the theoretical background of mixture designs, a mixture-process design has been established for 5 mixture components and 2 process variables. Scheffé type polynomial models have been developed for three measured responses (dE, viscosity and cost). These statistical models have been developed via the partial least squares regressions and analyzed through ANOVA and related diagnostic tools.Finally, models have been used for multi-response optimization by assigning goals for each response.Keywords: Mixture Designs; Mixture-Process designs; Experimental design for formulation; Industrial design of experiments; D-optimum criteria; Lean Six Sigma; DFSS; Multi-response optimization