A Multi-granular Linguistic Decision Model for Evaluating the Quality of Network Services

Resumen

The fuzzy linguistic approach has been applied successfully to many problems dealing with qualitative aspects that are assessed by means of linguistic terms. The use of linguistic information implies in most cases the need for using fusion processes to obtain aggregated values that summarize the input information. One important limitation of the fuzzy linguistic approach appears when fusion processes are applied to problems in which the linguistic information is assessed in linguistic term sets with different granularity of uncertainty, i.e., different cardinality; this type of information is denoted as multi-granular linguistic information. This limitation consists of the difficulty in dealing with this type of information in fusion processes due to the fact that there is no standard normalization process for this type of information, as in the numerical domain.In this contribution, taking as base the 2-tuple fuzzy linguistic representation model and its computational technique, we shall present a method for easily dealing with multi-granular linguistic information in fusion processes. Afterwards, we shall apply this fusion method to a decision process in a multi-expert decision-making (MEDM) problem with multi-granular linguistic information, that evaluates the quality of network services from different Operative Systems.