Generalised probabilistic linguistic evidential reasoning approach for multi-criteria decision-making under uncertainty. Issue 1 (2nd January 2021)

• Record Type: Journal Article
• Authors: Fang, Ran; Liao, Huchang; Yang, Jian-Bo; Xu, Dong-Ling
• Is Part Of: Journal of the Operational Research Society. Volume 72:Issue 1(2021)
• Page Start: 130
• Page End: 144
• Publication Date: 2021-01-02
• Subjects: General probabilistic linguistic term set -- evidential reasoning approach -- gained and lost dominance score method -- screening of high-risk population for lung cancer; Operations research -- Periodicals; 658.4034
• DOI: 10.1080/01605682.2019.1654415
• Languages: English
• ISSNs: 0160-5682

Abstract

Abstract: As a multi-criteria decision-making (MCDM) method, the evidential reasoning (ER) approach can deal with uncertainties that are resulted from the limited knowledge and experience of experts. Due to the lack of information in the decision-making process, experts usually cannot give quantitative evaluations, but can only express their views with linguistic terms. It is usually impossible for experts to give accurate linguistic terms since they may hesitate among several linguistic terms or interval ones. In addition, experts may have different preferences for different linguistic evaluations. To fully express the evaluations, the probability can be introduced to model the preferences of experts. In this study, we propose the generalised probabilistic linguistic term set (G-PLTS) to represent the evaluation information with various linguistic forms. Then, the ER approach is investigated in the environment with G-PLTSs. Besides, a gained and lost dominance score (GLDS) method is utilised to rank the alternatives, forming an integrated method, which we call the generalised probabilistic linguistic evidential reasoning (GPLER) approach, to solve the MCDM problems with several uncertainties. Finally, we apply this method to the screening of high-risk population of lung cancer to verify the effectiveness of the proposed method.