<p>Considering the characteristics of imprecise, incomplete and fuzzy data in emergency environment, a novel emergency decision-making method based on coverage-based variable precision (<i>I</i>, <i>PSO</i>)-fuzzy rough set model is proposed. First, an improved (<i>I</i>, <i>PSO</i>)-fuzzy rough set model is proposed, which combines the covering-based fuzzy rough set (CFRS) and the variable precision fuzzy rough set (VPFRS). Second, inspired by the idea of attribute reduction, a novel method for determining attribute weights is introduced to optimize weight assignment in emergency decision-making. Last but not least, to illustrate the feasibility and effectiveness of the proposed method, an example of post-flood rescue force allocation in urban areas is demonstrated. Finally, the stability and superiority of the method are verified through sensitivity analysis and comparative evaluation.</p>