A heterogeneous multi-attribute case retrieval method for emergency decision making based on bidirectional projection and TODIM. (1st October 2022)

• Record Type: Journal Article
• Authors: Zheng, Jing; Wang, Ying-Ming; Zhang, Kai; Gao, Jian-Qing; Yang, Long-Hao
• Is Part Of: Expert systems with applications. Volume 203(2022)
• Publication Date: 2022-10-01
• Subjects: Case retrieval -- Bidirectional projection -- TODIM -- Emergency decision making -- Heterogeneous multi-attribute; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2022.117382
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: Case retrieval method for EDM based on bidirectional projection and TODIM. DMs' behavior is considered for comprehensive utility of similar historical cases. The proposed method considers distance, angle, and bidirectional projection. The proposed method is valuable for heterogeneous multi-attribute decision-making. Abstract: Emergency decision making (EDM) entails a heterogeneous multi-attributes problem concerning crisp number, interval number, triangular fuzzy number, and intuitionistic fuzzy number, etc. However, it creates difficulties for the decision makers (DMs) in undertaking an optimal decision. The study focuses on the heterogeneous multi-attribute case retrieval method, which is an effective method to assist DMs in promptly generating an emergency alternative. Nevertheless, the accuracy and comprehensiveness of similarity measurement remains one of the major issues to be solved when obtaining the similar case set; this is a key step in EDM. A heterogeneous multi-attribute case similarity measurement based on bidirectional projection is firstly proposed to calculate the case similarities between the historical cases and the target case, which simultaneously considers distance and direction, along with the optimal model based on maximum entropy method and the nearness degree is constructed to determine the attribute weights. Furthermore, the DMs' psychological behavior is considered during the selection process of the most suitable historical case via the TODIM (an acronym in Portuguese of interactive and multi-attribute decision making) method. Hence, an alternative of the most suitable historical case can be exploited to generate the alternative of target case. Finally, the proposed method is illustrated using a typhoon disaster, and its feasibility, effectiveness, and practicality are validated via sensitivity analysis and its comparisons with other methods.