The analytic hierarchy process based on fuzzy simulation (FSAHP) is proposed in this paper. The weights of criteria and performance of alternatives are represented by the fuzzy linguistic variables, which are characterized as triangular fuzzy variables. Fuzzy simulation is used to estimate the expected values of fuzzy weights of all alternatives under final goal. The best alternative is the one whose expected value is biggest Finally, FSAHP is employed to evaluate three computer laboratories to illustrate its effectiveness