Aplicação de rede de filas multi-classes para a reestruturação de fluxos nos serviços hospitalares de urgência e emergência

Abstract

In Brazil, it has become common to find patients waiting for days in emergency services, especially public ones, representing a dramatic situation that has unfortunately been normalized. Emergency services struggle to admit patients, often turning into large inpatient wards. Strategies such as segmenting patients through risk classification—separating them into low, medium, and high clinical risk—and introducing three new areas for better flow management (fast-track, teleconsultation, and clinical decision units) are some of the strategies adopted to reduce waiting times. This dissertation, an action research, aims to promote the alignment between the installed capacity of resources (care facilities and healthcare professionals) in the hospital emergency department, treated here as an Open Queueing Network (OQN), with patient demand while ensuring service quality and patient safety. The model captures hospital-specific differences in the clinical risk profiles of patients, arrival patterns and volumes, and process efficiency within a single computational framework. Queueing equations are implemented to optimize the emergency department's areas using wait time, utilization factor, and overflow probability as service quality targets. Applying this methodology resulted in a significant reduction in wait times, enabling faster and more efficient responses to medical emergencies. Waste elimination and optimized patient flow not only enhanced operational efficiency but also substantially improved patient experience. The better-structured care team now delivers superior-quality service.