Optimization of Hospital Bed Occupancy in Hospitals Using Double Deep Q Network (DDQN)

Abstract

Obviously, the disappointments found in the emergency clinics the executives have typically identified with the absence of data and inadequate assets the board. The utilization of Deep Q Network (DDQN) can add to defeat these restrictions to distinguish applicable information on patient's administration and giving significant data to administrators to help their choices. All through this investigation were actuated DDQN models competent to make expectations in a real-time hospital environment and utilizing real clinical data. Considering this the proposed model is developed using the OpenAI Gym system, and show its utilization on a basic clinical bed occupancy model. Also, deep RL Agents utilizing PyTorch, and the Hospital Simulation is developed using SimPy. Research work exhibit a model utilizing a Double Deep Q Network as the Deep Reinforcement Learning Agent (DL Agent).