Instances of the article: "Data-driven risk-based maximal covering location problem with mobile units for healthcare emergency response"

Equitable access to healthcare during emergencies is critical yet challenging due to geographic and infrastructure constraints, particularly in developing nations. This study develops a novel two-phase hierarchical optimization model integrating data-driven risk assessment with facility location decisions to strategically deploy fixed hospitals and mobile healthcare units during health crises. The model combines multi-criteria and lexicographic optimization: Phase 1 identifies globally efficient resource configurations through weighted multi-criteria optimization considering coverage, accessibility, and redundancy; Phase 2 refines allocations by reallocating mobile units to maximize risk-weighted access prioritizing vulnerable populations. Municipal-level risk indicators are derived from vulnerability factors including comorbidities, demographics, and population density using machine learning techniques applied to individual COVID-19 patient records. Applied to Mexico (32 states, 2,478 municipalities, 126 million population, 638 hospitals), the model achieves high risk coverage in short computational time. Comparative analysis demonstrates that national-level coordination outperforms state-level independent planning in risk coverage. The approach demonstrates adaptability to diverse health emergencies beyond COVID-19, including seasonal outbreaks, epidemics, and future pandemics.