Service Selection Based on Customer Rating of Quality of Service Attributes

Abstract

Selecting the optimal service from a set of functionally equivalent services is non-trivial. Previous research has addressed this issue making use of Quality of Service (QoS) attributes of the candidate services. In doing this, researchers have however assumed that the customers' preference of the various QoS attributes varies linearly with the actual attribute values. In this work, we put forward a technique that overcomes this restriction and compares functionally equivalent services on the basis of the customers' perception of the QoS attributes rather than the actual attribute values. We utilize the `mid-level splitting' method to track the customer's preference vis-a-vis the actual attribute values. Further, we utilize the `Hypothetical Equivalents and In equivalents Method' to assign weights, reflecting the importance, to the attributes on the basis of the customer preference. The whole procedure is demonstrated using a simple running example.