The probabilistic linguistic term sets based ORESTE method for risk evaluation in Fine-Kinney model with interactive risk factors

Abstract

The Fine-Kinney model is a quantitative and effective method to identify and evaluate potential risks. The Fine-Kinney method has been widely used in practice, while the traditional Fine-Kinney method is difficult to access risk parameters precisely in practice. Besides, the current Fine-Kinney method fails to take into account the fact that decision makers are interrelated in practice. Further, the detailed relationships among the potential hazards cannot be reflected in the conventional Fine-Kinney method during the risk priority process, especially in the case of uncertain information. To compensate these deficiencies, this paper proposes an extended Fine-Kinney framework by integrating ORESTE (organísation, rangement et Syn-thèse de données relarionnelles) (in French), Choquet integral, and Probabilistic Linguistic Term Sets (PLTSs). Firstly, the PLTSs are utilized to express the decision makers’ complex risk preference information. Then, the Choquet integral is used to integrate risk evaluation information, which can simulate the potential interaction relationships among individual risk evaluation of decision-makers. Next, an extended ORESTE based on the PLTSs method is used to obtain the priorities of potential hazards, in which distance measure of PLTSs is applied to replace Besson’s ranks. Moreover, the PIR (preference, indifference, and incomparability) structure is constructed to describe the detailed relationships between potential hazards. Finally, an illustrative example is described to illustrate the proposed risk evaluation method. After that, the rationality and efficiency of the proposed method are tested through the comparison with other similar methods.