Neural Networks And Fuzzy Control for Bulk Terminal Operating Management
From Proceeding of Summer Computer Simulation Conference,SCSC95, Ottawa, July 24-26 1994
Pietro Giribone, Agostino G.Bruzzone

---

THE INTEGRATED DSS

The decision support system proposed creates the operating configuration illustrated in fig. 2. As can be seen, the information supplied by the simulation model is summarised in the fuzzy terms used to feed the neural model.
The use of a linguistic type of representation, such as the one that is possible using fuzzy logic, allows the manager to understand the operating conditions in which the DSS thinks it should operate. This by itself provides the manager with an important summary and allows him to obtain an immediate logical validation during the calibration phase and operating activities.
Similarly, the ANN outputs are filtered by a module that summarises them into operating decisions and always based on the variables with a linguistic connotation.
All this provides a high degree of understandability of the decision process, thus overcoming one of the fundamental limits in using ANN to develop DSS.