DOSE

The main causes of overcrowding of Emergency Operating Rooms (EORs) are the growing demand for non-elective surgeries and the unprecedented medical staff shortage. The effects of overcrowding can unfortunately ripple across the hospital, resulting in increasing in-hospital mortality, longer length of stays and higher costs. DOSE project is motivated by this context, where it becomes crucial to rethink the operations engineering and optimization of EORs. The objective of DOSE is to develop a multidisciplinary data-driven approach for optimizing EOR patient care pathways. More specifically, we aim at 1) developing data-driven approaches to provide accurate and comprehensive modeling of EOR patient care pathways, including predictions for resource requirements and activity delays; 2) developing health outcome Key Performance Indicators (KPIs) with machine learning techniques and medical expertise, a Quality of Working Life model for the EOR staff with human and social sciences, and a health-economic approach for cost-effectiveness analyses; 3) deriving optimal scheduling policies for care pathway activities by accounting for the above KPIs, and integrating the results into a digital twin to support real-time decision making. DOSE will improve EOR-related care delivery by exploiting tools from Industry 4.0 such as big data, data-driven optimization, and digital twins. It will be carried out in close collaboration with the hospital La Pitié Salpêtrière, and using detailed data on EOR patients from Entrepôt de Données de Santé. The project is challenging given the high dimension and the inherent stochastic and dynamic nature of the related problems, and our ambition to center on benefits and impacts for patients and medical staff. DOSE aims at bringing multidisciplinary contributions to the literature of data analytics, stochastic optimization, simulation, medicine, health economy, and human and social sciences.