4.5.3. Refinement of the solution
The use of the feedback mechanism described in Section 3.2.7 is shown in this section through a simple example. Consider the setting of Section 4.5.1, where the external conditions are assumed to be fixed, and a grid of 7 control options with step 25 ha from 1300 to 1450 ha is considered in the first DSS iteration. Weights w2 favouring the economic criterion are used in the multicriteria analysis. Observing Fig. 6 (solid line), it is apparent that the solution with the highest score among those evaluated is 1425 ha, but, due to the discrete approach used, the true optimum could be nearby at a level not actually studied. For this reason, the decision maker may choose to refine the solution by running another DSS iteration where the evaluated control options are in the interval [1400, 1450] ha with step 5 ha, while external conditions and multicriteria analysis weights are not changed. The AHP scores obtained in the second DSS iteration are shown in Fig. 7, and it can be observed that the solution corresponding to 1430 ha turns out to have an AHP score higher than 1425 ha. However, this does not mean that the decision maker would accept 1430 ha. From the analysis of Fig. 7, it is apparent that 1430 ha is an edge value after which the performance degrades quickly. Therefore, the decision maker might decide to keep the solution on lower values of aquaculture area (1425 ha or less) for a more robust decision with respect to uncertain factors not taken into account.