In a Mamdani fuzzy inference system (FIS), the rule weight represents the importance of a rule in making the overall decision. The rule weight is usually used during the aggregation stage to combine the output of each rule.
During the implementation stage, the degree of membership of the antecedent part is determined using the minimum method, which is the standard method used in Mamdani FIS. The product method can also be used, but it is less common.
During the aggregation stage, the rule weight is used to combine the consequent part of each fired rule. The standard method used in Mamdani FIS is the weighted average method, where each rule's consequent part is multiplied by its corresponding rule weight and then averaged with the consequent parts of the other fired rules.
The role of rule weight in a Mamdani FIS is important because it allows the designer to adjust the impact of each rule on the overall output. By adjusting the rule weights, the designer can fine-tune the system's behavior to better fit the desired output. However, it is worth noting that the rule weight does not affect the degree of membership of the antecedent part, which is determined solely by the input value and the membership function.
