Dynamic Case-Based Emergency Decision-Making Model under Time-Varying Single-Valued Neutrosophic Set

Jing Zheng,Ying-Ming Wang,Kai Zhang
DOI: https://doi.org/10.1016/j.eswa.2024.123830
IF: 8.5
2024-01-01
Expert Systems with Applications
Abstract:Case-based emergency decision-making (CEDM) models are effective tools for generating emergency response alternatives. Existing studies have indicated their merits in emergency response; however, representing and fusing ambiguous and dynamically changing case information remains a challenging issue. To address these problems, this study investigates a dynamic CEDM (DCEDM) model in a time-varying single-valued neutrosophic set (TVSNS) environment. First, TVSNS is defined to depict complex case information, which is not only fuzzy and uncertain but also continuously changing. Concurrently, the integral of the TVSNS is explored to aggregate the case information over multiple periods, which can preserve the integrity of case information in hesitation and ambiguous situations. Subsequently, a new attribute similarity measure under a neutrosophic set is developed. Additionally, an optimization model considering the interactive relationships among attributes is proposed to determine the attribute weights with Shapley values. Subsequently, a novel case similarity calculation method by incorporating the gained and lost dominance score and regret theory is explored, considering the regret avoidance behavior of decision-makers and fairness among attributes. The effectiveness of the proposed DCEDM model is verified with the generation of treatment alternatives for infectious COVID-19. Sensitivity and comparison analyses are provided to demonstrate the superiority of the proposed DCEDM model. The results reveal that the proposed DCEDM model can reflect the actual demand and better assist decision-makers in emergency response.
What problem does this paper attempt to address?