Scholarlink

<p>NSGA-II has been applied for multi-objective optimization, mostly proposing several dominant solutions as the Pareto front, resulting in uncertainty in the final decision. This paper contributes to this problem by proposing an ensemble multi-optimization algorithm combining Genetic Algorithm (GA) and Optimality Grade (OG), called GA-OG. This new GA-based algorithm replaces the solutions’ dominance in NSGA-II with the multi-objective optimality grade of solutions, resulting in a unique solution. The performance of GA-OG is evaluated against NSGA-II via solving a newly developed multi-objective, multi-period, mixed-integer linear mathematical model for viable healthcare resource planning. The proposed model minimizes the total costs, surplus beds, surplus specialists, and carbon footprint effects in a resilient cancer hospital’s location-allocation problem. The results of the model-solving in the case study of Iran using both GA-OG and NSGA-II revealed that GA-OG significantly outperforms NSGA-II in terms of computational efficiency while maintaining comparable solution quality. Specifically, GA-OG achieves a much lower runtime, while the optimality grade values remain nearly identical. In addition, unlike NSGA-II, which generates multiple Pareto solutions, even up to 36, GA-OG provides a single optimal solution, simplifying decision-making with less uncertainty. However, according to the results of the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), GA-OG, with a configuration of 200 iterations and a population of 50, proposes the best multi-objective optimization in the case study. Accordingly, a 10-year plan is proposed for cancer hospital location, cancer patient allocation, specialist recruitment, and part-time specialists’ allocation in the case study.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Sustainable-resilient healthcare resources planning using GA-OG: a new ensemble GA-based multi-objective optimization algorithm

  • 期刊论文

  • 发表日期：

    2025年10月15日

查看全文
  • Afshin Kordi,
  • Arash Nemati

摘要

NSGA-II has been applied for multi-objective optimization, mostly proposing several dominant solutions as the Pareto front, resulting in uncertainty in the final decision. This paper contributes to this problem by proposing an ensemble multi-optimization algorithm combining Genetic Algorithm (GA) and Optimality Grade (OG), called GA-OG. This new GA-based algorithm replaces the solutions’ dominance in NSGA-II with the multi-objective optimality grade of solutions, resulting in a unique solution. The performance of GA-OG is evaluated against NSGA-II via solving a newly developed multi-objective, multi-period, mixed-integer linear mathematical model for viable healthcare resource planning. The proposed model minimizes the total costs, surplus beds, surplus specialists, and carbon footprint effects in a resilient cancer hospital’s location-allocation problem. The results of the model-solving in the case study of Iran using both GA-OG and NSGA-II revealed that GA-OG significantly outperforms NSGA-II in terms of computational efficiency while maintaining comparable solution quality. Specifically, GA-OG achieves a much lower runtime, while the optimality grade values remain nearly identical. In addition, unlike NSGA-II, which generates multiple Pareto solutions, even up to 36, GA-OG provides a single optimal solution, simplifying decision-making with less uncertainty. However, according to the results of the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), GA-OG, with a configuration of 200 iterations and a population of 50, proposes the best multi-objective optimization in the case study. Accordingly, a 10-year plan is proposed for cancer hospital location, cancer patient allocation, specialist recruitment, and part-time specialists’ allocation in the case study.