Optimizing hospital resource allocation through demand forecasting in the Spanish health system

Abstract

This article presents an integrated model for budget optimization in healthcare systems, combining hospital demand forecasting techniques with linear programming for efficient resource allocation. The COVID-19 pandemic revealed significant limitations in the capacity of healthcare systems to respond to sudden increases in hospitalizations, highlighting the need for quantitative tools to support anticipatory planning. The proposed methodology is based on two complementary components. First, time series models are used to project short-term hospitalizations, capturing trends and seasonal patterns. Second, these projections are incorporated into a linear optimization model that determines the optimal number of hospital beds to be funded, taking into account budget constraints, unit costs, minimum proportions between general and intensive care beds, and basic operational requirements. The approach is validated through a case study using real data on COVID-19 hospitalizations in Spain obtained from official open sources. The results show that the projected demand remains at moderate and relatively stable levels, enabling the identification of economically viable and operationally efficient bed allocation configurations. Overall, the integration of predictive models and optimization tools proves to be a robust strategy for strengthening hospital planning in uncertain contexts. The proposed model facilitates informed, transparent, and justifiable decision-making, contributing to more efficient management of budgets and hospital capacity in contemporary healthcare systems.