Multi-objective student nurse allocation problem during training using exact and hybrid heuristic–metaheuristic methods
- 1Department of Industrial Engineering and Futures Studies, Faculty of Engineering, University of Isfahan
- 2Department of Nursing, Amol Faculty of Nursing and Midwifery, Mazandaran University of Medical Sciences
Abstract
Assigning nursing students to clinical training wards while satisfying operational, educational, and institutional constraints is a difficult multi-objective combinatorial optimization problem known as the student nurse allocation problem (SNAP). Based on real-world scheduling requirements gathered from expert interviews and university practice, this study develops a mixed-integer programming model for SNAP. The optimization criteria are minimizing the total completion time across wards and minimizing ward switching across courses. Two solution approaches are proposed: the augmented epsilon-constraint (AUGMECON) method for obtaining exact non-dominated solutions in small-size instances and a constructive-heuristic-guided NSGA-II algorithm (CH-NSGA-II) for larger instances. Computational experiments were conducted on generated instances with different sizes and on a real-world case study. In small-size instances, the AUGMECON and CH-NSGA-II were evaluated using computational time and distance-based indicator. For larger generated instances, CH-NSGA-II was compared with a randomized constructive heuristic (RCH) baseline using the normalized hypervolume indicator. The results show that CH-NSGA-II generated non-dominated solutions in shorter computational times than AUGMECON for the tested small-size instances. In the larger instances, CH-NSGA-II achieved higher normalized hypervolume values than RCH across all tested problem groups, indicating improved Pareto-front quality compared with the randomized constructive baseline. In the real-world case, the best reported solution reduced the total completion time across wards by 30.9% compared with the available manual schedule, while satisfying the considered hard constraints. The results suggest that CH-NSGA-II can support clinical placement scheduling in nursing education by providing alternative trade-off solutions for decision makers.
Related
- Effects of video reinforcement following face-to-face CPR/AED training on knowledge, attitude, and simulated performance among frontline community workers: a quasi-experimental study with post-training randomized allocation
- Teacher Training on Healthy Canteen Improves Knowledge and Attitude of Primary School Student
- Gamified Versus Traditional Balance Training: Effects on Balance and Motivation in Collegiate Student-Athletes
- Student perceptions of technology-enhanced simulation resources in laparoscopic training: a scoping review and implications for surgical instrumentation.
- Teacher-student-machine interaction autonomous learning: a structured LLM-integrated framework for developing independent clinical reasoning in residency training.
BibTeX
@article{afarani2026multiobjective,
title = {{Multi-objective student nurse allocation problem during training using exact and hybrid heuristic–metaheuristic methods}},
author = {Hanieh Shabani Afarani and Arezoo Atighehchian and Roghieh Nazari},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62702-0},
url = {https://doi.org/10.1038/s41598-026-62702-0},
note = {CC BY},
}
RIS
TY - JOUR
TI - Multi-objective student nurse allocation problem during training using exact and hybrid heuristic–metaheuristic methods
AU - Hanieh Shabani Afarani
AU - Arezoo Atighehchian
AU - Roghieh Nazari
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62702-0
UR - https://doi.org/10.1038/s41598-026-62702-0
AB - Assigning nursing students to clinical training wards while satisfying operational, educational, and institutional constraints is a difficult multi-objective combinatorial optimization problem known as the student nurse allocation problem (SNAP). Based on real-world scheduling requirements gathered from expert interviews and university practice, this study develops a mixed-integer programming model for SNAP. The optimization criteria are minimizing the total completion time across wards and minimizing ward switching across courses. Two solution approaches are proposed: the augmented epsilon-constraint (AUGMECON) method for obtaining exact non-dominated solutions in small-size instances and a constructive-heuristic-guided NSGA-II algorithm (CH-NSGA-II) for larger instances. Computational experiments were conducted on generated instances with different sizes and on a real-world case study. In small-size instances, the AUGMECON and CH-NSGA-II were evaluated using computational time and distance-based indicator. For larger generated instances, CH-NSGA-II was compared with a randomized constructive heuristic (RCH) baseline using the normalized hypervolume indicator. The results show that CH-NSGA-II generated non-dominated solutions in shorter computational times than AUGMECON for the tested small-size instances. In the larger instances, CH-NSGA-II achieved higher normalized hypervolume values than RCH across all tested problem groups, indicating improved Pareto-front quality compared with the randomized constructive baseline. In the real-world case, the best reported solution reduced the total completion time across wards by 30.9% compared with the available manual schedule, while satisfying the considered hard constraints. The results suggest that CH-NSGA-II can support clinical placement scheduling in nursing education by providing alternative trade-off solutions for decision makers.
SN - 2045-2322
ER -