Data for "A Compatibility-Aware Multi-Department Nurse Staffing-Block Assignment Model with Overtime and Cross-Unit Support"

Published: 29 April 2026| Version 1 | DOI: 10.17632/t2ckt8mds7.1
Contributor:
Min-Yang Li

Description

This dataset contains the synthetic computational results for a study of a compatibility-aware multi-department nurse staffing-block assignment model for hospital staffing policy analysis. The model integrates staffing cost, overtime, skill and grade requirements, weekly rest, policy-based cross-unit support eligibility, support premiums, and scenario-based compatibility penalties in a mixed-integer linear programming formulation. The computational study evaluates time-limited MILP benchmarks, a repair-based genetic algorithm, an ALNS-style exact-repair improvement stage, and a fix-and-optimize intensification stage. The package includes CSV outputs for the reported computational experiments: standalone and seeded algorithm benchmarks, hard-case per-run budget comparisons, ablation tests, staffing and compatibility sensitivity analyses, replicated policy-family comparisons, support-premium response analysis, daily department-consistency robustness checks, multi-seed downstream robustness, fix-and-optimize window sensitivity, and pipeline time-budget sensitivity. It also includes the manuscript figures derived from these outputs. All instances are synthetic. The compatibility profiles are scenario inputs rather than observed behavioral measurements. No patient data, hospital administrative data, nurse-level personnel records, or personally identifiable information are included.

Files

Institutions

Categories

Applied Sciences, Operations Research, Personnel Scheduling, Mixed Integer Programming

Licence