خلاصة:
Selection problems which contain many criteria are important and complex problems that involve different approaches have been proposed to fulfill this job. The Analytic Hierarchy Process (AHP) can be very useful in obtaining a likely result which can consider the decision maker’s subjective ideas. On the other hand, the Data Envelopment Analysis (DEA) has been a popular method for measuring the relative efficiency of decision making units (DMUs) and ranking them objectively in quantitative data. In this paper, a three-step procedure based on both DEA and AHP was formulated and applied to a case study. The procedure maintained the philosophy inherent in DEA by allowing each DMU to generate its own vector of weights. These vectors of weights were used to construct a group of pairwise comparison matrices which were perfectly consistent. Then, we utilized group AHP method to produce the best common weights compatible with the DMUs judgments. Using the proposed approach can give precise evaluation, combining the subjective opinion with the objective data of the relevant factors. The applicability of the proposed integrated model was illustrated using a real data set of a case study, which consisted of 19 facility layout alternatives.
ملخص الجهاز:
DEA is a non-parametric method for measuring the efficiency of a set of decision- making units (DMUs), such as firms or public sector agencies (Azadi, Jafarian, Farzipoor Saen and Mirhedayatian, 2015).
In other words, each DMU was asked (as a decision maker) to compare the relative importance of inputs/outputs and a pairwise comparison matrix was developed using the efficiency judgments (by solving one of the DEA models).
, 2013), utility functions (Brock, 1980; Keeney and Kirkwood, 1975; Greco, Kadziński, Mousseau and Słowiński, 2012; Huang, Chang, Li and Lin, 2013), and the AHP (Dyer and Forman, 1992; Van Den Honert and Lootsma, 1997; Chiclana, Herrera and Herrera-Viedma, 2001; Altuzarra Moreno-Jimenez and Salvador, 2010).
In the past three decades, multiple methods have been proposed to determine the weights of individuals (Ramanathan and Ganesh, 1994; Saaty, 1994; Forman and Peniwati, 1998; Bolloju, 2001; Van den Honert, 2001).
One important reason for such an approach is that the group Euclidean distance is a universal cardinal error measure which in many cases perfectly follows the purpose of the AHP to calculate cardinal information (weights) and not only ranks of alternatives like many other multi-criteria methods do (Azadeh, Ghaderi and Izadbakhsh, 2008; Alonso et al.
In other words, we ask each DMU (as a decision maker) to compare the relative importance of inputs/outputs and each DMU constructs its pairwise comparison judgment matrix based on the best weights produced according to a DEA model.
Aggregation of utility-based individual preferences for group decision-making, European Journal of Operational Research, Vol. 229, pp.