Deep Reinforcement Learning Approach for Emergency Response Management

The emergency medical services sector must take the responsibility of increasing the efficiency of the emergency response and reducing the travel time of the patient. However, most existing studies on emergency medical services (EMS) management are suffering from modeling complex emergency responses process To solve this issue. In his paper, we developed a reinforcement learning model free approach to deal with emergency response scheme for emergency response for emergency medical facilities. In practical terms, we first formulate the Markov decision process for EMS which present the characteristics of state, reward function and the action space. then, designed deep reinforcement learning to find the optimal behaviors for emergency responses management to reach the location of emergency calls and pick up the patients to hospital timely.