Dec 31, 1964
A mathematical model designed to help implement a major conclusion of RAND's continuing study of Air Force R&D policies (namely, that the most efficient way to conduct R&D may be to undertake parallel projects in the early and cheaper stages of a development program). The model is an initial attempt, through the use of sequential-decision theory, to explore parallel development in a more quantitative way. For example, how many conpeting projects should be run? And what are the factors that affect the answer to this question? The adoption of parallel-path strategy by the U.S. Air Force as an R&D policy seems most clearly indicated when the technical advances sought are large, when considerable uncertainty exists as to which of several development alternatives is the best to pursue, and when the cost of prototypes is small compared with the total system cost.