IMPLEMENTATION OF FRACTION OF DESIGN SPACE TO ASSESS PREDICTION CAPABILITY OF RESPONSE SURFACE DESIGN TO OPTIMIZE MODIFIED RELEASE TABLETS OF HYDROCHLOROTHIAZIDEAbstract
Selecting the best possible design is an uphill task considering a lot of criteria which must be scrutinized while choosing any design out of many alternative design options available. The experimental design is often compared and selected based on alphabetical design optimality criteria, like A, D, G or V. Each of them attempts to summarize any one important characteristic of the selected design. G-optimality criteria favor those experimental designs which are having the smallest maximum prediction variance over the entire region of interest. It does not consider the distribution of the magnitude of the prediction variance throughout the region. Graphical methods like variance dispersion graph and the fraction of design space plots can be used to overcome this drawback. These graphical methods are explored extensively for comparing the quality of prediction throughout the design space. It serves as a useful tool to assess the prediction performance of the design and making a comparison between experimental designs. The fraction of the design space plot provides a comprehensive picture of the predictive quality of experimental design through the entire region of interest. Single curve provides detailed information of scaled prediction variance for an assumed model and specific design region for the selected design. Here Fraction of design space plots is adapted to evaluate designs for optimization of modified release tablets of hydrochlorothiazide to get desired in-vitro dissolution profile. Three response surface methodology designs are compared by applying this plot including Central Composite Circumscribed and Face centered design along with Box-Behnken design using Design expert software for modeling.
V. D. Singhai *, S. Sharma and S. Paliwal
Alembic Research Center, Alembic Pharmaceuticals Limited, Vadodara, Gujarat, India.
24 December 2018
29 April 2019
14 August 2019
01 September 2019