摘要
随着中国人口老龄化的加剧,居家养老服务需求迅速增加。在一些大型社区中,由于对居家养老业务的需求量庞大,护工的调度问题变得愈发重要。但目前仅有少量文献对此进行研究,且都是基于已知护工数量的前提,无法为服务中心招收各等级护工数量提供参考。本文提出一种基于均值漂移聚类和遗传算法的护工调度算法,在完成所有老人护理任务的前提下进行护工的合理调配。算法首先将老人按照位置及所需护工等级进行聚类,以减少计算规模;然后采用遗传算法,使用实数编码方式和精英保留策略,对护工的服务路径进行规划,提高效率;最后将文中提出的算法应用于实际数据。实验结果表明,对老人进行聚类后再采用遗传算法能有效地降低成本、减少路径规划所需的时间。通过分析结果中所需不同等级的护工数量,给出了所需各等级护工数量与服务老人数量之间比例关系的建议。
With the intensification of China’s aging population,home health care services have been required increasingly.In some large communities,due to the huge demand for home health care services,the scheduling of nursing workers has been regarded as something more important.However,this issue is only addressed in a few studies based on the premise of the known number of nursing workers,through which a reference for the service center to recruit the number of nursing workers of different levels cannot be provided.A caregiver scheduling algorithm,which is based on mean shift clustering and genetic algorithm,is proposed in this paper.And reasonable allocation of caregivers is made under the condition of completing all the elderly nursing tasks.In this algorithm,the elderly are clustered first according to the required caregiver grade and position to reduce the calculation scale;then genetic algorithms,real-number coding and elite retention strategies are all used to plan the service path of the caregiver to improve efficiency.Finally,the algorithm proposed in this paper is applied to real data.The results that the cost and the time required for path planning after clustering the elderly can be reduced effectively by the genetic algorithm is shown in this experiment.Through the analysis of the number of different levels of nursing workers required in the results,some suggestions are provided on the proportional relationship between the number of nursing workers required in each level and the number of the elderly served.
作者
胡竣涛
时小虎
马德印
HU Juntao;SHI Xiaohu;MA Deyin(School of Computer Science and Engineering,Changchun University of Technology,Changchun Jilin 130012,China;College of Computer Science and Technology,Jilin University,Changchun Jilin 130012,China;Business School,Jilin University,Changchun Jilin 130012,China)
出处
《广西师范大学学报（自然科学版）》
CAS
北大核心
2021年第3期27-39,共13页
Journal of Guangxi Normal University:Natural Science Edition
基金
国家自然科学基金(61972174)
吉林省教育厅社会科学项目(JJKH20190210SK)
吉林省科技厅科技发展计划项目(20200101146FG)。
关键词
均值漂移
遗传算法
护工调度
多路径规划
mean shift
genetic algorithm
nursing workers scheduling
multipath planning
作者简介
通信作者:马德印(1978-),男,长春工业大学讲师,博士。E-mail:madeyin@mail.ccut.edu.cn。
