In order to overcome the defects in the traditional TOPSIS model of river health assessment, such as correlations between indices, uncertainty in estimating the weights subjectively, difficulty in determining the membership function and possibility of the scheme closed to ideal solution and negative-ideal solution concurrently,an improved algorithm was presented. At first, the decision-making matrix based on standard samples and evaluation objects was built.Then,the weight of every evaluation index was decided by using AHP and PPC method, and the weighting decision matrix was constructed.At last,Euclid distance was replaced with Chi-square distance to calculate the proximity degree of each alternative to ideal solution. On the basis of this, the evaluation object was graded by the approaching degree of the object and the standard stylebook sequence.In this way,the non-determinacy and randomness resulting from the subjective judgments in the assessment,as well as the problem that coinstantaneous scheme may be ideal or not,were avoided,and the scientificity and rationality of TOPSIS Model can be improved. At the end of this paper,the feasibility and validity of this method were testified by the actual problem.
