An application of multi-objective simulation optimization to medical resource allocation for the emergency department in Taiwan

• Published in: Annals of operations research Vol. 326; no. 1; pp. 199 - 221
• Main Authors: Kuo, R. J., Song, P. F., Nguyen, Thi Phuong Quyen, Yang, T. J.
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2023; Springer; Springer Nature B.V
• Subjects: Algorithms; Business and Management; Combinatorics; Emergency medical care; Emergency medical services; Emergency service; Evolutionary computation; Genetic algorithms; Heuristic methods; Hospitals; Mathematical optimization; Methods; Multiple objective analysis; Operations research; Operations Research/Decision Theory; Optimization; Original Research; Overcrowding; Resource allocation; Simulation; Simulation models; Sorting algorithms; Theory of Computation; Utilization; Taiwan; Non-dominated sorting differential evolution algorithm (NSDE); Emergency department; Non-dominated sorting genetic algorithm (NSGA II); Non-dominated sorting particle swarm algorithm (NSPSO); Simulation optimization
• ISSN: 0254-5330; 1572-9338
• DOI: 10.1007/s10479-023-05374-7

In recent years, the need for emergency resources has dramatically increased and it has caused an overcrowding problem for the emergency department (ED). Solving this problem by increasing medical resources is either impractical or infeasible. Thus, this manuscript develops a multi-objective mathematical model to allocate medical resources for the emergency department (ED). The optimal resource allocation is exploited by using some meta-heuristic algorithms, i.e., fast and elitism non-dominated sorting genetic algorithm (NSGA II), non-dominated sorting particle swarm algorithm (NSPSO), and non-dominated sorting differential evolution (NSDE). Thereafter, a dynamic simulation model, which embeds the solutions from the resource allocation model in the simulation process, is constructed. Each feasible solution from the three meta-heuristic algorithms is simulated to estimate the performance of the resources allocated in terms of the average service level and staff utilization. The results show that the performance of the NSGAII, where the average service level and staff utilization for the current resources are 0.844 and 0.751, respectively, is better than those of NSPSO and NSDE. Besides, the number of medical staff gives a significant effect on the service level and utilization while the number of beds only impacts staff utilization. The simulation model can find out that the best combination of the number of staff and the number of beds is from 1 to 10 staffs and 1–6 beds to maximize the utilization and service level.