Statistical Design and Analysis of Experiments

A problem solving strategy employing experimental designs is illustrated in optimizing the yield of a secondary metabolite in solid substrate culture. The strategy employs a screening Plackett-Burman design that selects two of six factors. Then a two- level factorial design is carried out in order to define a search direction. The optimizing search is done employing the Simplex algorithm. A surface response quadratic model in three factors is generated to predict optimum conditions.