The Economics of Parallel R and D Efforts: A Sequential-Decision Analysis

An examination of the logic of the parallel-path approach to the solution of development problems. If there is considerable uncertainty about the best way to achieve a particular objective, and if it is possible to reduce uncertainty substantially in the early stages of a development effort, parallel work on several alternatives may be a very good strategy. It may even be good strategy to develop completely several systems, test them for performance, calculate their production cost, and then let a production contract for the system that has the best combination of performance and (non R and D) cost. If time is important, a good development strategy should reflect this fact both in running more projects in parallel, and by spending at a faster rate on each of the projects undertaken.