الفهرس | Only 14 pages are availabe for public view |
Abstract The recent directions for solving multi-criteria decision-making (MCDM) problems are mainly depending upon the achieved progress in computer science technologies such as: expert system, decision support system and artificial neural network (ANN). These technologies have been applied to overcome the shortcoming associated with the solution of MCDM problems. In this thesis, an ANN has been developed and trained to rank the set of criteria by considering the decision-maker preferences. The network topology, training process, advantages and disadvantages are discussed. In addition, the interaction style between the decision-maker (DM) and the network is clarified. Also we build a knowledge base depending on the trained ANN The developed ANN has been applied to solve an engineering application which zs a multi-objective linear programming problem (MOLPP) in a large steel producing unit located in the city of Annaba, in Algeria. |