Selecting the best portfolio alternative from a hybrid multiobjective GA-MCDM approach for New Product Development in the pharmaceutical industry

摘要

A fundamental challenge in managing a pharmaceutical company is identifying the optimal allocation of finite resources across the infinite constellation of available investment opportunities. In that context, the optimal management of the new product pipeline has emerged at the forefront of all strategic planning initiatives of a company. The combined use of discrete-event simulation and multi-objective optimization methods to minimize both the failure risk of the product, development time while maximizing profits (Net Present Value - NPV) was identified as an efficient framework in our previous works. In that context, multiobjective Genetic Algorithms are particularly attractive for treating this kind of problem, due to their ability to directly lead to the so-called Pareto front, from which a single solution has to be chosen by the decision maker. To help this final decisional process, this work proposes to resort to multicriteria decision making methods such as ELECTRE, PROMETHEE, TOPSIS and also a new and simple method called FUCA to select the best alternative. The three criteria considered are NPV, risk and makespan for a New Product Development (NPD) problem in the pharmaceutical industry. The four methods are compared on a test-bench example from the dedicated literature, and the conclusion that one would be expect is that no method overcomes the others in any situation. TOPSIS is an all-purpose method giving sometimes worst results, while ELECTRE and PROMETHEE are more efficient when the decision maker preferences are crisply defined. For the proposed example, the FUCA procedure shows a good efficiency.