Reinforcement effect evaluation for dangerous sluices is a complicated system, in which the evaluation index information, index weight and evaluation set are all fuzzy and uncertain. In view of this, according to the Administrative measures for the post evaluation of reinforcement projects by the central government, a comprehensive evaluation index system, which consists of five first-level indexes and 21 second-level indexes, is established for the reinforcement effect of dangerous sluices, and the grading standards under the hundred percentage point system is determined. Furthermore, cloud models for index weight, evaluation set and the representation of evaluation information are established, and the numerical characteristics of such cloud models are used to quantify the propagation of uncertainty in the evaluation process and the credibility and stability of evaluation results. The reinforcement project of Taipu sluice is studied as an example. Research results show that the uncertainty of index weight has great influence on evaluation results, and the proposed method is effective and can be used for similar projects. The reinforcement effect of Taipu sluice can be determined as “supreme” level, which accords with the actual situation.