By arranging the assessment indices into the independent levels, i.e., the level of basic factors, including the characteristics of the drainage area, the reservoir morphology and the hydrological features, the level of hydrochemical factors, and the level of biological factors, and determining the effective indices in each level through a Delphi investigation, the system of indices, complete and simple, was established.

Reservoirs were classified into five ranks according to their fish productivity (from high level to low level). On the basis of the results of the Delphi invetigation, the criterion of every index corresponding to each rank was determined. Thereby, the system of index criterion was worked out.

The importance of different indices in indicating fish productivity is quite different, so, a different weight should be given to each of them. The index weight system was .established by Delphi investigation and Grey Statistical Method.

In view of the fuzziness of the distinction between high and low fish productivity, Fuzzy Method was used to establish a comprehensive assessment model. First, Fuzzy Membership Function of each index to fish productivity was determined according to the index criterion system, which was comparable to each other on a same scale and can be used as a specific fish productivity index (FPL_i). Then all the FPL_i in the same level was added up according to the index weight system, thus a comprehensive fish productivity index (FPL_c) for each level was obtained. Finally, an overall conclusion of reservoir fish productivity was drawn by means of averaging FPL_c of all three levels, o by refined classification.