COMPARISON OF COSTS FOR MULTI-STAGE GROUP TESTING METHODS IN THE PHARMACEUTICAL INDUSTRY

Abstract

The issue of whether to test components individually, or alternatively in groups, in order to detect certain chemical properties remains an important issue in the pharmaceutical industry. Economic viability is of paramount importance since, for multi-stage procedures, the cost of additional stages must be taken into consideration, along with the cost of testing mixtures of components. Optimum groups sizes are calculated for the two-stage, three-stage (both members of Li's family of algorithms) and the row-and-column procedures. The γ-two-stage design is investigated, which involves using a γ-separating design at the first stage, followed (if necessary) by a strongly separating design at the second stage. Finally, comparisons are made between the costs of single and multi-stage procedures, for both optimum and standard groups sizes, through the use of two different cost functions.