在發生到院前心跳停止（OHCA）事件時，於黃金救援時間內提供緊急醫療救護對於最大化患者生存機率至關重要。然而，由於緊急醫療服務（EMS）資源有限，即時的電擊去顫（AED）或心肺復甦術（CPR）常難以實現。為解決此問題，本研究探討偕同註冊急救先鋒APP的志願者，於EMS抵達前先行提供即時救護。本研究旨在優化自動體外心臟除顫器（AED）的佈署位置與志願者派遣策略，以最大化每一個OHCA患者的期望生存機率，並確保在考量志願者隨機行為的情況下，OHCA患者在黃金救援時間內獲得急救的比例達到一指定的服務水準。
本研究提出一個考慮隨機限制式的二元變數混整數模擬最佳化（Mixed Binary and Integer Simulation Optimization）模型求解Pre-EMS資源配置問題，並與國家災害防救科技中心（National Science and Technology Center for Disaster Reduction, NCDR）合作，透過NCDR提供之真實的歷史OHCA資料、歷史人流資料及真實路網資料，採用數據驅動的方法構建隨機模擬模型。最後以高雄市三民區為研究地區，驗證所提出之演算法的效能，並為OHCA緊急醫療服務提供具有價值的決策建議。

Out-of-hospital cardiac arrest (OHCA) requires emergency medical care within the golden rescue time to maximize the patient’s chance of survival. However, limited Emergency Medical Services (EMS) resources often hinder the timely delivery of defibrillation (AED) or cardiopulmonary resuscitation (CPR). To address this challenge, this study explores the involvement of volunteers registered with the "First on Scene" app to provide immediate pre-EMS care. The objective is to optimize the placement of Automated External Defibrillators (AEDs) and the dispatch strategy of volunteers, in order to maximize the expected survival rate of OHCA patients while ensuring that a specified proportion receive emergency care within the golden time, even under the uncertainty of volunteer behavior.
We propose a data-driven mixed binary and integer simulation-optimization model with stochastic constraints to determine pre-EMS resource allocation. In collaboration with the National Science and Technology Center for Disaster Reduction (NCDR) in Taiwan, we incorporate real-world OHCA incident records, pedestrian flow data, and road network information to construct a realistic simulation framework. The model is solved using a hybrid approach that combines Rapid Screening (RS) and Nested Partitioning (NP) with an OCBA-CO-based resource allocation strategy. We apply the proposed method to Kaohsiung City’s Sanmin District to evaluate its effectiveness and to offer actionable insights for enhancing OHCA emergency response systems.