Selecting a portfolio of nonhomogeneous R \& D proposals (Q1060951)

This article reports a procedure developed to assist the U.S. Department of Energy in the selection of a portfolio of proposals addressing the design and manufacture of components of solar photovoltaic arrays, and complete photovoltaic systems. Considerations such as the relative importance of the different components in the overall effort, compatibility of the proposed designs, and budgetary and programmatic issues were complicating factors. The technical quality for each proposal was represented by multiple evaluation measures. These were combined into a single index through the use of a multiattribute utility function. Indices representing the relative importance of the various components were then developed, and results in measurable value theory were applied to derive an index of technical quality of a portfolio of proposals. The problem was then formulated as an integer program. It was solved by using an interactive branch and bound code that allowed for sensitivity analysis before selecting the portfolio of proposals for funding. The procedure was successfully applied to funding a portfolio of proposals aimed at reducing the cost of producing photovoltaic arrays, and to several other government-sponsored procurements.