A new fuzzy operator and its application to topology design of distributed local area networks

Topology design of distributed local area networks can be classified as a hard combinatorial optimization problem. The problem has several conflicting objectives, such as cost, reliability, network delay, and number of hops between source and destination. These desirable features of a network topology can best be expressed in linguistic terms, which is the basis of fuzzy logic. This paper presents an approach based on fuzzy logic that combines the conflicting objectives into a single optimization function. A new fuzzy operator, namely UAO is also proposed, and a structured approach based on fuzzy rules and preference rules is introduced. The UAO operator is empirically compared with another well-known operator called OWA through application in an evolutionary algorithm. Results suggest that the UAO exhibits comparatively better performance than the OWA operator.