At the end of the inference process, the system provides a map where several fuzzy set characterize each geographical object. The defuzzyfication module transforms fuzzy sets map in a map of ``potentialities'' c, that is, a real value between 0 and 1 where 1 denotes the best solution and 0 the worse.

We have chosen a method to reinforce the importance of the core of the final fuzzy set which contains the main information [10].

For example, assume n fuzzy sets expressed by characterizing a point (x,y). The crisp value c(x,y) is calculated by: