Abstract: Bids evaluation quality determines the degree of compatibility between bidders and a bidding project, however under the influence of experiential knowledge and subjective judgment, the quantifying process of qualitative indexes tend to be of subjective fuzziness, which could affect results accuracy. In order to lessen the adverse effects of fuzziness on bids evaluation results, Z-number fuzzy number is introduced to recognize the uncertainty with qualitative indexes; in accordance with entropy principle, bids evaluation criteria and the corresponding weights assigned by evaluation experts can be determined; the TOPSIS technique is used to integrate the index values, an improved TOPSIS method of bids evaluation model can then be established based on Z-number theory. The results show that the proposed method can determine the fuzziness of the qualitative indexes and realizes a reduction in the uncertainty with index values. Meanwhile, on the basis of avoiding the uncertainty due to subjective weights determination, the proposed method also overcomes the incommensurability between indexes. By conducting example and sensitivity analysis on proposed model, its effectiveness and stability are verified.