Abstract
This work proposes a hybrid approach based on rough set methodology and fuzzy inference system, for extraction of classification rules. Rough set methodology was used to eliminate abundant information of chemical indices while maintaining a lesser degree of information loss from the raw data; then the relationships among the variables under consideration were created using the fuzzy inference system which takes into consideration the experience of the decision-maker during fitting the membership functions. An application to assessment of groundwater quality is also given. Results demonstrate that the proposed approach has the ability to deal with the nonlinearity of the variables, as well as it deal with the highly subjective nature of the variables with highest degree of efficiency and so it can be used for finding the solution of multiple criteria decision making problems.