工业工程

工业工程 ›› 2024, Vol. 27 ›› Issue (5): 33-42.doi: 10.3969/j.issn.1007-7375.240210

• 服务运营管理与调度优化 • 上一篇    

考虑重症监护资源的手术集成调度

杨帆1, 聂广孟2, 王克霞2, 李慧1, 谭嘉文3, 谢小磊4   

  1. 1. 清华大学 医院管理研究院，广东 深圳 518055;
    2. 清华大学附属北京清华长庚医院，北京 102218;
    3. 中山大学附属第一医院，广东 广州 510062;
    4. 清华大学 工业工程系，北京 100084
  • 收稿日期:2024-05-20 发布日期:2024-11-05
  • 通讯作者: 谢小磊(1986—)，男，河南省人，副教授，博士，主要研究方向为医院运营管理、电子病历数据挖掘和基于系统科学的政策分析。Email: xxie@tsinghua.edu.cn E-mail:xxie@tsinghua.edu.cn
  • 作者简介:杨帆(2000—)，男，福建省人，硕士研究生，主要研究方向为医疗调度优化。Email: yangfanb2581@163.com
  • 基金资助:
    国家自然科学基金资助项目(72271137)；北京市医院管理中心青年人才培养“青苗”计划资助项目(QML20210904)

Integrated Surgical Scheduling Considering Intensive Care Resource

YANG Fan1, NIE Guangmeng2, WANG Kexia2, LI Hui1, TAN Jiawen3, XIE Xiaolei4   

  1. 1. Institute for Hospital Management, Tsinghua University, Shenzhen 518055, China;
    2. Beijing Tsinghua Changgung Hospital, Beijing 102218, China;
    3. The First Affiliated Hospital, Sun Yat-sen University, Guangzhou 510062, China;
    4. Department of Industrial Engineering, Tsinghua University, Beijing 100084, China
  • Received:2024-05-20 Published:2024-11-05

PDF

3

摘要： 手术集成调度是指在调度过程中对手术患者所需的手术室、手术人员和病床等各类资源加以协调。缺少配备监护设备以及医护人员的ICU病房床位等重症监护资源将给医疗系统带来严峻的挑战，如手术延迟、术后健康恶化等。使用传统手术调度模型所得方案忽略了重症监护资源的运行压力，基于此，本文建立一个旨在同时优化手术室效率和重症监护室运行压力的双目标函数规划模型。为有效求解这一模型，本文提出一种改进的遗传算法，通过向变异算子引入多样化的策略增强探索的灵活性，同时提出了修复算子以提高复杂约束下解的可行性。采用仿真驱动方法生成初始种群的一部分个体，有效克服大规模问题求解中常见的难以生成可行初始解的困难。基于真实医院数据的实验表明，在小规模算例下，算法获得接近最优解，在50名患者的规模下，相比于仅考虑手术室的手术调度策略，算法能够降低10.00% ~ 28.56%的重症监护室最大运行压力。

关键词: 服务调度, 改进遗传算法, 集成调度, 医院运营管理

Abstract: Integrated surgical scheduling refers to the coordination of various resources required for surgical patients, such as operating rooms, surgical staff, and hospital beds, during the scheduling process. The lack of intensive care resources, such as ICU beds equipped with monitoring equipment and staffed with medical personnel, poses a severe challenge to the healthcare system, such as delayed surgeries and postoperative health deterioration. The policies obtained using traditional surgical scheduling models often ignore the operating pressure of ICU resources. A bi-objective programming model was developed to simultaneously optimize the efficiency of the operating room and the operating pressure of the ICU. A modified genetic algorithm was proposed that enhances the flexibility of exploration by introducing diverse strategies to the mutation operator, and a repair operator was introduced to improve the feasibility of solutions under complex constraints. By employing a simulation-driven initialization approach, this study effectively overcomes the difficulty of generating a valid initial population in solving large-scale problems. Experiments based on real hospital data show that the algorithm obtains near-optimal solutions for small-scale arithmetic cases, and at the scale of 50 patients, the algorithm is able to reduce the maximum operating pressure of ICU by 10.00% to 28.56% compared to the surgery scheduling policy that only considers the operating room.

Key words: service scheduling, modified genetic algorithm, integrated scheduling, hospital operations management

中图分类号: