Based on the fuzzy probability theory of the third kind of fuzzy events, a new multi-objective optimization model of fuzzy probability is developed. And the application procedure of the fuzzy probability model is illuminated in detail. In order to assess index weighted value scientifically, the analytic hierarchy process (AHP) is introduced into the fuzzy weighted values to avoid the uncertainties while determining the index weights. In this way, the fuzzy probability model not only holds the merit of classical fuzzy comprehensive evaluation method, but also overcomes its limitations in practice. The model can utilize the known fuzzy objective information sufficiently and can also satisfy the requirements of decision maker as much as possible. In fact, if the fuzzy weights are not taken into account in the fuzzy probability model, it will degenerate to the classical fuzzy comprehensive evaluation model. The fuzzy probability model is based on clear concept and rigorous theory, and can be applied extensively to obtain objective and dependable result for multiple attribute decision making problems.