یک روش جدید برای مقابله با عدم قطعیت تصمیم گیرندگان در فرآیند انتخاب تجهیزات

One of the main problems faced while configuring or reconfiguring manufacturing systems is to rank alternative designs taking into account all the different aspects involved (both tangible and intangible). For this purpose the Analytic Hierarchy Process (AHP) is a well-known decision making support method that addresses this problem. A major drawback of AHP is that uncertainty in the judgments of the decision makers and the resulting impact on the ranking is not considered. In real situations, however, judgments based on perceived future scenarios are almost always uncertain. To solve this problem in this paper we present the first complete probabilistic extension to the AHP method. The new method provides the decision maker not only with information on the ranking of the alternatives but also the probability that the ranking remains stable even in presence of uncertainty in the judgements. We verified the validity of the new method in a real application developed for the Ferrari racing team.

Multi criteria decision making is a very complex problem in modern industry, as clearly stated by Krause and Golm [ I ] and by Elwany, Khairy, Abou-Ali and Harraz [2]. Analytic Hierarchy Process is a decision making support instrument, developed by T. L. Saaty in 1980 [3], useful to for managing multi criteria decisions. AHP has been widely used in industry in cases of investment selection, make or buy decisions, supplier selection, and configuration or reconfiguration of manufacturing systems. The method consists in tackling the selection among alternatives, a complex multidimensional decisional problem, by using a hierarchy of comparisons based on simpler criteria. At each level the decision maker defines the weight of any alternative referred to a single criterion and then he evaluates the relative importance of the criteria in the context of the decision at hand. AHP allows the tackling of problems taking into account both its tangible and intangible aspects. AHP also allows the formalization of the decisional process, in order to better justify the decisions and later to find the causes of possible mistakes. Therefore AHP is also a company knowledge management tool and favours the professional growth of the people involved in its use. At any level of the decisional tree pairwise comparisons are used to give the weights to the criteria within the hierarchy and to the alternatives at the bottom of it. Each judgement defines the relative importance of two items when compared referring to the higher level criterion in the hierarchy.