Mamdani Inference
A fuzzy-inference approach in which rule consequents are fuzzy sets whose activated outputs are aggregated and commonly defuzzified into a crisp value.
In Mamdani inference, the consequent of a rule is itself expressed as a fuzzy set.
Example:
IF temperature is high AND humidity is high THEN fan_speed is highInput membership grades activate rules to different degrees. The resulting fuzzy consequents are aggregated, and a defuzzification step is commonly used when a crisp numerical output is required.
The approach is useful when expert knowledge should remain visible as readable rules. Its cost appears as the rule base grows: maintenance becomes harder, and membership-function tuning can substantially change the controller response.
Mamdani and Sugeno inference are not interchangeable descriptions. Sugeno consequents are typically constants or functions of the inputs rather than output fuzzy sets.
Examples appear in Fuzzy Logic.