ORAHS2025
Abstract Submission

147. Dynamic Predictive Modeling for Decision Support and Patient Management in Intensive Care Units

Invited abstract in session MD-3: ED and ICU, stream Sessions.

Monday, 13:30-15:00
Room: NTNU, Realfagbygget R9

Authors (first author is the speaker)

1. Daniel García-Vicuña
Department of Statistic, IT and Mathematics, Public University of Navarre
2. Ana María Anaya-Arenas
Université du Québec à Montréal
3. Janosch Ortmann
GERAD, CRM and UQAM
4. Angel Ruiz
Department of Operations and Decision Systems, Universite Laval
5. Fermin Mallor
Institute of Smart Cities, Public University of Navarre

Abstract

The limited availability of Intensive Care Unit (ICU) beds highlights the critical challenges hospitals face when demand exceeds capacity. In such contexts, healthcare providers must make complex decisions regarding admissions, discharges, and surgery scheduling—often under significant uncertainty and pressure. These decisions, while necessary, can lead to logistical bottlenecks and adverse patient outcomes. This work presents a dynamic predictive model that supports decision-making by anticipating ICU patient trajectories and enabling more efficient bed management. Built on real-world patient data, the model combines clinical variables and historical information to estimate ICU length of stay and evolving occupancy levels. The approach integrates predictive analytics with simulation techniques to evaluate the operational impact of different strategies under variable demand scenarios. Our results show that informed forecasting can help reduce premature discharges and surgical delays by guiding clinicians toward more balanced resource allocation. The model is designed as a practical decision-support tool to assist ICU managers in aligning patient care needs with available capacity, particularly in periods of high stress or crisis.

Keywords

Status: accepted


Back to the list of papers