This paper presents an optimal weighting approach for maximizing the overall preference value of decision alternatives based on a given set of weights and performance ratings. In policy analysis settings, relative weights for policy alternatives are subjectively assessed by a group of experts or stakeholders via surveys using comparative judgment. A hierarchical pairwise comparison process is developed to help make comparative judgment among a large number of alternatives with fuzzy ratio values. Performance ratings for policy alternatives are obtained from objective measurement or subjective judgement. The preference value of an expert on a policy alternative is obtained by multiplying the weight of the alternative by its performance rating. Two optimization models are developed to determine the optimal weights that maximize the overall preference value of all experts or stakeholders. An empirical study of evaluating Taiwan's air cargo development strategies is conducted to illustrate the approach.