A Simulation-based Analysis Using Machine Learning Models to Optimize Patient Flow and Treatment Costs

Abstract

This study investigates the application of discrete event simulation in analyzing patient management and cost dynamics within a hospital system. A simulation that integrates machine learning models, specifically Decision Trees, Random Forests, Support Vector Machines, and Gradient Boosting methods, to predict treatment costs and appointment availability was developed. Conducted over 30 days, the simulation generates synthetic data for training the models. The results are assessed in terms of the total number of patients treated, cumulative costs incurred, and the cost-effectiveness of each predictive model. The findings reveal significant variations in the performance of different machine learning techniques, demonstrating that adopting advanced analytics can substantially improve hospital resource management. This research aims to develop more efficient patient care strategies, contributing to optimizing hospital operations and enhancing patient experiences.