Empathic network learning for multi-expert emergency decision-making under incomplete and inconsistent information

TL;DR

This paper presents a novel method for learning empathic networks among decision-makers in emergencies, handling incomplete and inconsistent preference data to improve decision quality.

Contribution

It introduces an empathic network learning approach based on robust ordinal regression and preference disaggregation, addressing data incompleteness and inconsistency in emergency decision-making.

Findings

Successfully completed incomplete fuzzy judgment matrices

Calculated intrinsic utilities of decision-makers

Derived representative empathic networks for emergency scenarios

Abstract

Challenges, such as a lack of information for emergency decision-making, time pressure, and limited knowledge of experts acting as decision-makers (DMs), can result in the generation of poor or inconsistent indirect information regarding DMs' preferences. Simultaneously, the empathic relationship represents a tangible social connection within the context of actual emergency decision-making, with the structure of the empathic network being a significant factor influencing the outcomes of the decision-making process. To deduce the empathic network underpinning the decision behaviors of DMs from incomplete and inconsistent preference information, we introduce an empathic network learning methodology rooted in the concept of robust ordinal regression via preference disaggregation. Firstly, we complete incomplete fuzzy judgment matrices including holistic preference information given in terms of decision examples on some reference alternatives, independently by each DM, and we calculate the intrinsic utilities of DMs. Secondly, we establish constraints for empathic network learning models based on empathic preference information and information about relations between some reference nodes. Then, the necessary and possible empathic relationships between any two DMs are calculated. Lastly, tailored to the specific requirements of different emergency scenarios, we design six target networks and construct models to derive the most representative empathic network.