An emergency decision-making method based on D-S evidence theory for probabilistic linguistic term sets. (July 2019)

• Record Type: Journal Article
• Authors: Li, Peng; Wei, Cuiping
• Is Part Of: International journal of disaster risk reduction. Volume 37(2019)
• Publication Date: 2019-07
• Subjects: Emergency decision-making -- D-S evidence theory -- Probabilistic linguistic term set -- Operational law; Emergency management -- Periodicals; Risk management -- Periodicals; Disaster relief -- Periodicals; Hazard mitigation -- Periodicals; 363.34
• DOI: 10.1016/j.ijdrr.2019.101178
• Languages: English
• ISSNs: 2212-4209

Abstract

Abstract: Emergency decision-making processes, always require decision makers (DMs) to express the majority of their opinions in linguistic terms (LTs). Since there are, in most cases, many more people than pure DMs participating in the decision-making process, probabilistic linguistic term sets (PLTSs) are suitable to express the decision information. A typical feature of emergency decision-making is that even minor errors in the decision process may have dire consequences. Accurate decision information and the consequent handling of that information are absolutely essential. We show that there are significant drawbacks in the existing operational laws for PLTSs, which may result in faulty decisions. In order to address the issue, in this paper, we propose some new operational laws for PLTSs based on D-S evidence theory and discuss the properties of these new operational laws and advance a probabilistic linguistic weighted averaging operator based on D-S evidence theory (DS-PLWA). Our aggregating method can keep the form of PLTS and avoid information loss. In addition, we propose a novel comparison method for PLTSs and a method to obtain the criteria weights based on maximising deviation for the evidences. On this basis, we propose an emergency decision-making style based on D-S evidence theory and apply it to a known details of a serious mine accident that happened in Pingyi, Shandong province.