A two-phase methodology for technology selection and system design

A two-phase methodology to guide research and development managers in the evaluation and selection of competing technologies is presented. Deterministic multi-attribute utility theory is used in the first phase to rank the technological alternatives and to eliminate inferior candidates. The procedure is illustrated with an application drawn from a study centering on the evaluation of electric and hybrid passenger vehicles. Thirty-nine individuals were interviewed to assess their risk preferences and determine their attitudes toward the vehicle design. In the second phase, it is assumed that a particular technology has been chosen for further development. The decision-maker must then allocate a fixed amount of resources to different projects, some of which may be undertaken in parallel, to maximize a given measure of performance. The problem is formulated as a probabilistic network and solved heuristically using Monte Carlo simulation. Results are presented for the most preferred vehicle identified in phase one for two representative decision-makers and three budget options. In each case, the heuristic finds a solution that corresponds to the optimal allocation of funds