Multiobjective R and D portfolio analysis in highly uncertain environments

The author addresses the modeling of resource allocation problems for R and D projects having multiple, competing objectives in an uncertain environment. The model developed (called NMOS) is different from an earlier general multiobjective formulation of the problem; the project selection environment is extended from the linear multiobjective case to a nonlinear multiobjective stochastic environment. It is shown that the inclusion of stochastic variables in the model leads to the concept of the level of objective functions and the probability of achieving those levels. This is achieved through the use of a technique called the PROTRADE (probabilistic tradeoff development) method. The application of the model is discussed and illustrated