An approach to emergency decision making based on decision-theoretic rough set over two universes

Abstract

Emergency decision making is still an important issue in unconventional emergency management research. Although many studies have been written on this topic, they remain political and qualitative, and it is difficult to make them operational in practice. Therefore, this article proposes a decision-theoretic rough set over two universes as an approach for solving this difficulty. The proposed approach integrates rough set theory over two universes using a Bayesian decision-making technique. In this study, emergency decision making is considered as a multiple-criteria ranking or multiple-criteria selection problem with multi-granularity linguistic assessment information. A Bayesian decision process based on linguistic description with qualitative data over two universes is first presented to construct the decision model and approach, and then the decision-theoretic rough set theory over two universes is taken to induce a set of decision rules that satisfy minimum risk of loss conditions. These rules can easily give the optimal decision results with minimum risk of loss by considering online information, realistic scenarios, and the dynamic characteristic of unconventional emergency events as they develop. Finally, the steps and the basic principle of the proposed method are illustrated by a numerical example with the background of emergency decision making.