A Two-Stage Adjustable Robust Optimization Model for Integrated Allocation of Beds, Medical Staff, Medicines and Equipment under Uncertain Patient Arrivals and Limited Capacity

Hospitals must commit beds, staff, medicines and equipment before they know how many patients will arrive, and a plan built on average demand may fail exactly when demand surges. This paper develops a two-stage adjustable robust optimization model for the integrated allocation of these four resources under uncertain, class-specific patient arrivals and limited capacity. First-stage capacities are fixed before the planning horizon; second-stage admission, overtime and emergency-procurement decisions respond to arrivals observed up to each day and are restricted to non-anticipative linear decision rules. Arrival uncertainty is described by a budgeted polyhedral set whose budget Γ controls the number of classes that may simultaneously reach their worst case in a day. The robust counterpart is an exact linear program. Four formal results are established: dualisation of the budgeted constraints, feasibility and attainment of the optimum, monotonicity in the budget, and a Hoeffding-based calibration rule that also shows when a probabilistic guarantee is uninformative. In a synthetic three-class hospital instance, the robust counterpart was verified on 3,000 sampled scenarios (maximum constraint violation 1.8 × 10⁻¹⁵) and compared with a nominal model and a sample average approximation on 2,000 out-of-sample scenarios under a non-anticipative admission policy. With the budget Γ = 1.5, the rejection rate fell from 10.8% to 2.7% and the tail cost (CVaR at 95%) by 3.9%, at a 6.2% higher mean cost. When the diversion penalty is doubled the robust design is 7.3% cheaper than the nominal one, and under a shifted demand law with a doubled penalty it outperforms the sample average approximation in mean (1.6%) and tail (5.6%) cost.

Authors

Institutions

Publication Details

Published
2026-11-21
DOI
https://doi.org/10.54117/cj5chj75
Primary Topic
Healthcare Operations and Scheduling Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Two-Stage Adjustable Robust Optimization Model for Integrated Allocation of Beds, Medical Staff, Medicines and Equipment under Uncertain Patient Arrivals and Limited Capacity

Henry E. Egbogho, Napoleon Ifakachukwu Ochonogor
Healthcare Operations and Scheduling Optimization
article

A Two-Stage Adjustable Robust Optimization Model for Integrated Allocation of Beds, Medical Staff, Medicines and Equipment under Uncertain Patient Arrivals and Limited Capacity

Henry E. Egbogho, Napoleon Ifakachukwu Ochonogor
article en

Abstract

Hospitals must commit beds, staff, medicines and equipment before they know how many patients will arrive, and a plan built on average demand may fail exactly when demand surges. This paper develops a two-stage adjustable robust optimization model for the integrated allocation of these four resources under uncertain, class-specific patient arrivals and limited capacity. First-stage capacities are fixed before the planning horizon; second-stage admission, overtime and emergency-procurement decisions respond to arrivals observed up to each day and are restricted to non-anticipative linear decision rules. Arrival uncertainty is described by a budgeted polyhedral set whose budget Γ controls the number of classes that may simultaneously reach their worst case in a day. The robust counterpart is an exact linear program. Four formal results are established: dualisation of the budgeted constraints, feasibility and attainment of the optimum, monotonicity in the budget, and a Hoeffding-based calibration rule that also shows when a probabilistic guarantee is uninformative. In a synthetic three-class hospital instance, the robust counterpart was verified on 3,000 sampled scenarios (maximum constraint violation 1.8 × 10⁻¹⁵) and compared with a nominal model and a sample average approximation on 2,000 out-of-sample scenarios under a non-anticipative admission policy. With the budget Γ = 1.5, the rejection rate fell from 10.8% to 2.7% and the tail cost (CVaR at 95%) by 3.9%, at a 6.2% higher mean cost. When the diversion penalty is doubled the robust design is 7.3% cheaper than the nominal one, and under a shifted demand law with a doubled penalty it outperforms the sample average approximation in mean (1.6%) and tail (5.6%) cost.

Vol. 1(1)
Dennis Osadebay University (NG)
Healthcare Operations and Scheduling Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon