Decision-making method based on set pair analysis and VIKOR under heterogeneous information environment and application to typhoon disaster assessment

• Published in: Soft computing (Berlin, Germany) Vol. 27; no. 12; pp. 8289 - 8314
• Main Authors: Tan, Ruipu, Zhang, Wende, Yang, Lehua
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2023
• Subjects: Application of Soft Computing; Artificial Intelligence; Computational Intelligence; Control; Engineering; Mathematical Logic and Foundations; Mechatronics; Robotics; Connection number; VIKOR method; Neutrosophic set; Set pair analysis; Multiple-attribute decision; Heterogeneous information
• ISSN: 1432-7643; 1433-7479
• DOI: 10.1007/s00500-022-07750-0

A multi-attribute decision-making method based on heterogeneous information and set pair analysis is proposed, considering the balance of the contradictions between information representation and information processing in the decision-making process. The method is applied to decision-making problems in which the attribute weights are completely unknown and the attribute values are characterized as heterogeneous information (including exact numbers, interval numbers, intuitionistic fuzzy numbers, interval intuitionistic fuzzy numbers, neutrosophic numbers, interval neutrosophic numbers, picture fuzzy numbers, and Pythagorean fuzzy numbers). First, various transformation methods for converting heterogeneous information into the connection numbers (CNs) of set pair analysis are introduced and defined, and the heterogeneous information is uniformly converted into CNs for ease of processing. Second, the attribute weights of the heterogeneous information are obtained using the information entropy of the CNs. Thereafter, the alternatives are ranked by the VIKOR method. Moreover, we use the method of setting a threshold for the compromise evaluation value, and the clustering method to group the alternatives to provide more auxiliary decision making. Finally, the feasibility and effectiveness of the proposed method are demonstrated by a typhoon disaster evaluation example, and the advantages of the proposed approach are emphasized by performing a multi-faceted comparative analysis with other methods. Among them, the time complexity of the proposed algorithm can be simplified as O m , which demonstrates an improvement on the simplified time complexity O m 2 of the two existing methods. Particularly in the context of big data, the advantages are apparent.