Statistical Experimental Design and its Application to Pharmaceutical Development Problems

Abstract

Pharmaceutical scientists are often confronted with the problem of developing formulations and processes for difficult products and must do so in spite of competing objectives. Pressures, placed on the scientist to balance variables and meet these objectives, can be compounded when limited funds, time and resources require rapid and accurate development activities. Statistical experimental designs provide an economical way to efficiently gain the most information while expending the least amount of experimental effort Some commonly used techniques called Response Surface Designs are presented. These include the Central Composite, Extreme Vertices, Simplex and Evolutionary Operations designs. Procedures for planning experiments and for analyzing and interpreting data are also discussed. Guidance is provided on the implementation of experimental results and the practical use of these techniques for routine experiments