The problem of objective ranking: foundations, approaches and applications

Abstract

The paper starts with the discussion of the issue of objectivity versus subjectivity, stressing that while an absolute objectivity is not attainable, nevertheless trying to be as objective as possible constitutes a higher value, necessary for hard science and technology. Dangers and errors of the subjectivist reduction of objectivity to power and money attempted by the postmodern sociology of science are discussed. Then we turn to the problem of subjective versus objective decision analysis and ranking. It is shown that while all classical decision theory aims at a rational analysis and support of subjective decisions, there are important application cases, particularly in managerial problems, when the decision maker prefers to avoid specifying her/his preferences and needs decision analysis – e.g., ranking of decision options – that is as objective as possible. An approach to decision support that might be easily adapted for such objective ranking is the reference point methodology; its application is shown on examples. One of these examples is actually not an application of the methodology, but a real life problem that motivated the development of objective ranking. The examples illustrate that objective ranking might be important for management, including also management of telecommunication networks.