Dealing with fuzzy costs in telecommunications networks design

Planners of telecommunications networks frequently are faced with uncertainties about the data values to be utilized. Conventional mathematical models using such data can therefore lead to weak results often dependent upon these uncertainties. An option is to model the problems by using techniques involving fuzzy sets and fuzzy numbers. Supported by these techniques, we develop a model that helps planning a stage in the telecommunications access networks. The model accepts uncertainties about equipment costs and permits analyzing competition aspects among different technologies and topologies. The mathematical problem to be solved is of mixed linear type with 0-1 variables. We also present an application to an actual network, focusing on the use of SONET/SDH unidirectional rings to serve key customers