Adaptive multi-source information fusion for intelligent decision-making in emergencies: Integrating personal, event, and environmental factors

This study proposes an innovative adaptive multi-source information fusion framework for intelligent decision-making in emergencies, integrating personal, event, and environmental factors. The framework introduces a novel two-layer factor indicator system, which adapts dynamically to evolving emergency conditions by adjusting the consideration of factors such as risk tolerance, cognitive load, event severity, event uncertainty, time urgency, and resource availability. A key innovation is the use of adaptive support degree and confidence metrics, enabling decision-makers to express their judgments and assess rapidly changing situations in a more structured and flexible manner. The framework further employs an adaptive hierarchical fusion strategy based on Dempster-Shafer theory, enabling the dynamic synthesis of multi-source information and the resolution of conflicting evidence in real-time. For group decision-making, a consistency-based expert fusion mechanism is introduced to ensure the harmonization of expert inputs, improving the robustness of decisions even under uncertainty. The proposed method is validated through a real-world disaster case in China, demonstrating its practical applicability, adaptability, and robustness. Additionally, a sensitivity analysis evaluates the performance of the framework under varying conditions, highlighting its ability to handle complex and dynamic emergency scenarios.