ATLAS: A Novel Approach to Generating Efficient D-Optimal Experiments for Acceleration of Biopharmaceutical Research & Development
Open AccessThe current proposal is a mixed method of research, comparing the efficacy of designed experiments and one-factor-at-a-time (OFAT) experiments in drug development within the biopharmaceutical industry. This study explores classical design techniques which normally ignore the interaction of the variables, confounding, as well as the aspect of optimization. The variables are complex and have multiple levels such as three, four, and even five. Data analysis was performed using multiple statistical functions such as the analysis of variance (ANOVA), correlation, and regression. This is aimed at establishing the associations between variables and other statistical trends. The programing language R has been used for development. The theoretical approach of Design of Experiments (DOE) and mathematical formulation of D-optimality and G-efficiency have been used to build the program. The ATLAS software developed as part of this research has the capability to predict a reduced number of optimal runs. This study surveys two data sets on which the ATLAS software is tested against traditionally designed experiments, such as, OFAT, full-factorial, and fractional-factorial designs. The research has demonstrated the use of DOE approach and D-optimality to provide novel optimization methods. A key finding from the study is that the ATLAS generated designs yield as much statistical insight, and in several cases more, than the classical designs in a reduced number of runs.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.