Table 2.
Description of the operation of the inference system used to calculate the Technical Risk.
| Point | Step | Description |
|---|---|---|
| 1 | Definition of the fuzzy sets and the membership functions See Figure 6-Step 1 |
Firstly, the different fuzzy sets and the respective membership functions for each input variable of the system are established [62,63,64,65]. These functions make it possible to determine the degree of membership of a certain value to a specific set in the range from zero to one. This is equivalent, respectively, to not belonging at all, or fully belonging to, the qualitative trait that represents the section of the membership function. Trapezoidal functions have been chosen for this, since they make it possible to maximize the degree of belonging to a set within a continuous range of values. |
| 2 | Fuzzification of input variables See Figure 6-Step 2 |
The fuzzification process of the input variables makes possible to determine the degree of membership (in the range from zero to one) of a specific value, associated with a certain input variable, within a certain fuzzy set. |
| 3 | Determination of the rules See Figure 6-Step 3 |
After the fuzzy sets and the membership functions have been determined, and the fuzzification process of the input variables has been carried out, then the rules governing the behavior of the system are established. These rules allow to perform the combination of the antecedents, that is of the input values, by using fuzzy operators of the ‘AND’ and ‘OR’ types. After this, combinations of the ‘IF’… ‘AND’/‘OR’… ‘THEN’ … types are carried out, allowing to connect the input values with the consequents, i.e., the outputs of the system. |
| 4 | Evaluation and application of the rules See Figure 6-Step 4 |
After performing the previous steps described in this table, it is possible to apply the fuzzy operators. There are different methods for their application [62]. The ‘AND’ operator will be used with the ‘intersection’ fuzzy operator and the ‘OR’ operator with the ‘union’ one, also equivalent to the ‘minimum’ and ‘maximum’ operators, respectively. When the ‘AND’ connector is used to connect two membership functions, the minimum of their membership degrees is returned, while in the case of the operator ‘OR’ the maximum one is returned. |
| 5 | Obtaining the output fuzzy set See Figure 6-Step 5 |
After evaluating and applying the rules, the consequent of these rules is determined by using the implication method [62]. The minimum is chosen in this case, which is translated into a truncation of the consequent’s membership function, that has been previously defined as a fuzzy set based on the value obtained from the fuzzy operation. |
| 6 | Aggregation of consequents See Figure 6-Step 6 |
After evaluating all the rules and obtaining their individual consequents, the global consequent (representative of the risk) is determined. |
| 7 | Defuzzification See Figure 6-Step 7 |
Finally, a numerical value for the risk level is obtained by applying the centroid method [62,66,67]. |