Many engineering systems face various natural and manmade hazards, whose occurrences can significantly disrupt system operations. The task of assessing how such systems perform under multiple hazards becomes especially challenging when hazards recur stochastically and influence one another in a dynamic manner. In this paper, we propose an analytical approach to model multiple dependent hazards, and use this model to quantify a system’s multimodal performance (expressed as availability) and its resilience. Specifically, we derive the joint distribution for the hazards’ occurrence and severity over time, then link that distribution to system degradation and recovery processes, ultimately determining the system’s multimodal availability. By applying this method to the IEEE 9-bus modified test system as an approximation of the Western System Coordinating Council (WSCC) in the United States, we illustrate both the effectiveness and the practical relevance of our proposed approach to system performance evaluation.