EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Hybrid Particle‐Firefly Adaptive Evolutionary Algorithm for Multiobjective Healthcare Scheduling: Minimizing Waiting Time, Travel Distance, and Workload Imbalance

Md. Limonur Rahman Lingkon, Md. Mohibul Islam, Md. Mushiur Rahman

Complexity · 2026-01 · 2026 (1)

DOI: 10.1155/cplx/4871498 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Complexity, online erschienen am 14.11.2026. Verfasst von Md. Limonur Rahman Lingkon und 2 weiteren Autor:innen. Die Arbeit nennt 74 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Diagnostic centers must assign patients’ tests to testing rooms (gates) and sequence them while three goals conflict: short and clinically fair waiting, little patient movement, and balanced room workloads. This study formulates a linear multiobjective mixed‐integer model that minimizes severity‐weighted total waiting time, mobility‐weighted movement, and capacity‐normalized workload imbalance subject to time windows, test precedence with buffers, gate–test compatibility, gate capacities and opening hours, and a high‐severity access constraint. A random‐key representation with a time‐window‐repairing decoder places seven multiobjective algorithms on the same search space: a new hybrid particle‐firefly adaptive evolutionary algorithm (HPFAEA), which combines particle‐swarm exploration, firefly refinement towards archive elites, evolutionary recombination, a pheromone memory on gate assignments and success‐based parameter adaptation, and multiobjective PSO, ACO, NSGA‐II, FA, ABC, and BFO. All algorithms were tuned with the Taguchi method and compared at equal numbers of function evaluations on 12 instances generated from real data and on the recorded day of a diagnostic center in Rajshahi, Bangladesh (50 patients, 272 tests, and 20 gates). Exact Pareto sets obtained with the augmented ε‐constraint method could only be certified for the smallest instances (up to 16 tests), where HPFAEA found on average 46.8% of the exact Pareto points within a fraction of a second; the remaining points were approached but not reached, quantifying the optimality gap. Overall instances, the tuned multiobjective ACO and PSO were strong competitors: HPFAEA ranked second by hypervolume (Friedman test p < 0.001), statistically tied with ACO and significantly better than the other five algorithms, and it obtained the best hypervolume on the largest instances and on the real case study. The ablation study shows that evolutionary recombination is the main source of the hybrid’s performance, whereas the firefly refinement step is counterproductive under a fixed evaluation budget. For the case study, the HPFAEA compromise schedule reduced severity‐weighted waiting by 93.1% and workload imbalance by 72.3% relative to the center’s first‐come‐first‐served, nearest‐room practice and removed all time‐window violations, at the cost of more patient walking, which the Pareto set makes explicit. The sensitivity analysis identifies the match between arrival times and room opening hours and the equipment capacity as the most influential model parameters.

Zitieren

APA

Lingkon, M. L. R., Islam, M. M., & Rahman, M. M. (2026). A Hybrid Particle‐Firefly Adaptive Evolutionary Algorithm for Multiobjective Healthcare Scheduling: Minimizing Waiting Time, Travel Distance, and Workload Imbalance. Complexity, 2026(1). https://doi.org/10.1155/cplx/4871498

Vancouver

Lingkon MLR, Islam MM, Rahman MM. A Hybrid Particle‐Firefly Adaptive Evolutionary Algorithm for Multiobjective Healthcare Scheduling: Minimizing Waiting Time, Travel Distance, and Workload Imbalance. Complexity. 2026;2026(1). doi:10.1155/cplx/4871498

BibTeX
@article{Lingkon2026,
  author = {Lingkon, Md. Limonur Rahman and Islam, Md. Mohibul and Rahman, Md. Mushiur},
  title = {A Hybrid Particle‐Firefly Adaptive Evolutionary Algorithm for Multiobjective Healthcare Scheduling: Minimizing Waiting Time, Travel Distance, and Workload Imbalance},
  journal = {Complexity},
  year = {2026},
  volume = {2026},
  number = {1},
  doi = {10.1155/cplx/4871498}
}
Kostenlos starten →