<p>Optimizing the locations of emergency facilities and mitigating risk are critical challenges in managing emergency logistics for hazardous materials (hazmat). Traditional research has often assumed deterministic travel times for emergency planning, but real-world scenarios frequently involve uncertainties in travel duration. This study addresses the hazmat emergency facility location and allocation problem by incorporating random emergency response times. We propose a distributionally robust optimization (DRO) model, a novel approach that explicitly accounts for response time uncertainty by optimizing under worst-case probability distributions within an ambiguity set. Unlike conventional deterministic and robust optimization methods, our DRO framework leverages limited historical data to construct ambiguity sets, allowing for more flexible and data-driven decision-making. This innovation significantly enhances the reliability and adaptability of emergency response planning by mitigating the risks associated with uncertain response times. Utilizing two methodologies to transform historical response time data into tractable ambiguity sets, we demonstrate the superior efficacy of our approach through a real-world case study in China. Experimental evaluations highlight the DRO model’s efficiency in balancing cost minimization and risk mitigation, outperforming traditional methods under uncertain conditions. Additionally, sensitivity analyses illustrate how key parameters influence system performance, offering critical insights for decision-makers in hazmat emergency management. The proposed methodology provides a robust decision-support tool that improves resilience in emergency logistics while ensuring computational efficiency.</p>