基于多目标改进免疫算法和GIS的养老机构空间配置优化研究——以上海市虹口区为例

doi:10.13249/j.cnki.sgs.2018.12.013

摘要/Abstract

摘要：

综合考虑政府、居民、投资者3方需求,构建带约束多目标的养老机构配置优化模型,基于改进免疫算法和GIS技术,对上海市虹口区养老机构的配置优化问题进行研究,分析研究区现有养老机构在空间分布和规模配置上的合理性,提出优化配置方案。研究表明：研究区现有养老机构数量缺口较大、部分养老机构偏离最佳区位、规模与需求存在较大差距;3所位于江湾镇街道的养老机构在现有区位运行欠合理;为充分满足居民养老需求,需在虹口区南部地区增设15所养老机构;通过与一般免疫算法、遗传算法、粒子群算法、模拟退火算法得到的优化结果对比可知,改进免疫算法在此优化问题中的求解效率分别提高45%,38.89%,21.43%,46.34%,求解精度分别提高1.61%,2.73%,5.80%,6.91%。

关键词: 养老机构, 配置优化, 改进免疫算法, GIS, 多目标

Abstract:

In recent years, the population is ageing rapidly in Shanghai. Increases in the older population are generating demand for a wide range of elder care services. Residential care homes are the main places for elderly care service. The configuration rationality of residential care homes makes a difference to sustainable urban development and the construction of a harmonious society. However, some issues in the configuration of residential care facilities such as unreasonable spatial layout and insufficient service have become increasingly acute with the growth of the geriatric population. It is important to optimize the layout and number of residential care facilities to ensure the equity and efficiency of public services. In this paper, a method for optimizing the layout and quantity of residential care homes together is studied. Considering the demand of government, residents and investors, a configuration optimization model with multi-objective and constraints for residential care homes is constructed based on maximizing the government equity, maximizing configuration efficiency, minimizing residents traveling cost and maximizing the investors’ returns. Then, the software GIS and the improved immune algorithm are used for the configuration optimization of the residential care homes in Hongkou District of Shanghai. According to the calculating results, the configuration rationality of current residential care homes is analyzed and a new configuration scheme is put forward. After comparing the results of the proposed method with those of simulated annealing algorithm, particle swarm algorithm, genetic algorithm, traditional immune algorithm and the existing scheme, the feasibility and superiority of the proposed method is verified validly. The results show that the scale and numbers of residential care homes still cannot satisfy the demand of the elderly in Hongkou District. Furthermore, most of the current residential care homes in Hongkou District deviate from the optimal locations. They may increase the residents' travel cost, weaken the government equity and configuration efficiency, hinder the increase of investment returns and need to be further optimized. According to the optimize results, the rationality of 3 residential care homes that are located in Jiangwanzhen Street should be further improved and 15 new residential care homes should be built in the south of Hongkou District to satisfy the demands of the elderly population. In addition, the computing efficiency of the improved immune algorithm is higher than that of simulated annealing algorithm, particle swarm algorithm, genetic algorithm, basic immune algorithm by 45%, 38.89%, 21.43%, 46.34% respectively, and the antibody affinity value of improved immune algorithm is better than that of the simulated annealing algorithm, particle swarm algorithm, genetic algorithm, basic immune algorithm respectively by 1.61%, 2.73%, 5.80%, 6.91%. The proposed method improves the basic immune algorithm from two aspects including selection operator and mutation operator. Hence, the computation procedure is more efficient and accurate than that of traditional immune algorithm. Due to considering varies stakeholders’ requirements, the configuration scheme is reasonable. The method can support knowledge-based policy-making and planning of residential care facilities. The optimized results also can provide references for scientific decision-making on residential care in Hongkou District of Shanghai.

Key words: residential care homes, configuration optimization, improved immune algorithm, GIS, multi-objective

中图分类号:

F299.24

程敏, 崔晓. 基于多目标改进免疫算法和GIS的养老机构空间配置优化研究——以上海市虹口区为例[J]. 地理科学, 2018, 38(12): 2049-2057.

Min Cheng, Xiao Cui. Spatial Optimization Configuration of the Residential Care Homes Based on the Multi-objective Improved Immune Algorithm and GIS: A Case Study of Hongkou District in Shanghai[J]. SCIENTIA GEOGRAPHICA SINICA, 2018, 38(12): 2049-2057.

图/表 8

图1

表1

图2

图3

表2

表3

图4

表4

参考文献 23

[1]	孙雅玮. 基于老年人口增长的上海市养老设施规划研究[J]. 世界地理研究, 2016, 25(5): 121-130. doi: 10.3969/j.issn.1004-9479.2016.05.013
	[Sun Yawei.Aged care facility planning of metropolitan Shanghai based on growth of the elderly population. World Regional Studies, 2016, 25(5): 121-130.] doi: 10.3969/j.issn.1004-9479.2016.05.013
[2]	许昕, 赵媛, 张新林, 等. 江苏省人口老龄化空间分异演变及影响因素[J]. 地理科学, 2017, 37(12): 1859-1866. doi: 10.13249/j.cnki.sgs.2017.12.009
	[Xu Xin, Zhao Yuan, Zhang Xinlin et al. Spatial variation of population aging and associated factors in Jiangsu Province. Scientia Geographica Sinica, 2017, 37(12): 1859-1866.] doi: 10.13249/j.cnki.sgs.2017.12.009
[3]	Thomas B.Factors related to the acceptance of home care and nursing homes among older Egyptians: A cross-sectional study[J]. International Journal of Nursing Studies, 2009, 46(5): 1585-1594. doi: 10.1016/j.ijnurstu.2009.05.006 pmid: 19500789
[4]	Chang Y,Schneider J K,Sessanna L.Decisional conflict among Chinese family caregivers regarding nursing home placement of older adults with dementia[J]. Journal of Aging Studies, 2011, 25(4): 436-444. doi: 10.1016/j.jaging.2011.05.001
[5]	刘红. 中国机构养老需求与供给分析[J]. 人口与经济, 2009, 25(4): 59-64.
	[Liu Hong.An analysis on demand and supply of institutional old aged caring in China. Population & Economics, 2009, 25(4): 59-64.]
[6]	Rioux L, Werner C.Residential satisfaction among aging people living in place[J]. Journal of Environmental Psychology, 2011, 31(2):158-169. doi: 10.1016/j.jenvp.2010.12.001
[7]	Cheng Y, Wang J, Rosenberg M W.Spatial access to residential care resources in Beijing, China[J]. International Journal of Health Geographics, 2012,11(1): 32-43. doi: 10.1186/1476-072X-11-32 pmid: 22877360
[8]	季翔,肖炳科,孙强. 城市机构养老服务设施规划配置布局研究——以徐州市主城区为例[J]. 现代城市研究, 2017, 32(2): 27-32. doi: 10.3969/j.issn.1009-6000.2017.02.004
	[Ji Xiang, Xiao Bingke, Sun Qiang.The research of planning and configuration layout strategy on urban institution endowment service facilities: A case study of main city zone of Xuzhou. Modern Urban Research, 2017, 32(2): 27-32.] doi: 10.3969/j.issn.1009-6000.2017.02.004
[9]	李灵芝,张嘉澍,仇白羽,等. 大规模保障住区养老设施配置优化研究——以南京市为例[J]. 现代城市研究, 2016, 31(6): 11-15. doi: 10.3969/j.issn.1009-6000.2016.06.002
	[Li Lingzhi, Zhang Jiashu, Qiu Baiyu et al. Allocation optimization for elderly facilities in public housing community: A case study in Nanjing city. Modern Urban Research, 2016, 31(6): 11-15.] doi: 10.3969/j.issn.1009-6000.2016.06.002
[10]	宋姗,王德,朱玮,等. 基于需求偏好的上海市养老机构空间配置研究[J]. 城市规划, 2016, 40(8): 77-82. doi: 10.11819/cpr20160813a
	[Song Shan, Wang De, Zhu Wei et al. Spatial allocation study on Shanghai’s nursing facilities for the elderly based on the elderly’s choice behavior. City Planning Review, 2016, 40(8): 77-82.] doi: 10.11819/cpr20160813a
[11]	Tao Z,Cheng Y,Dai T et al. Spatial optimization of residential care facility locations in Beijing, China: Maximum equity in accessibility[J].International Journal of Health Geographics, 2014,13(1): 33-44. doi: 10.1186/1476-072X-13-33 pmid: 4160561
[12]	王海鹰, 秦奋, 张新长, 等. 基于蚁群优化算法的城市生态用地空间规划模型[J]. 地理科学, 2017, 37(3): 426-436.
	[Wang Haiying, Qin Fen, Zhang Xinchang et al. The spatial planning model of urban ecological land based on ant colony optimization algorithm. Scientia Geographica Sinica, 2017, 37(3): 426-436.]
[13]	孙百生, 郭翠恩, 杨依天, 等. 基于GIS的承德乡村地名文化景观空间分布特征[J]. 地理科学, 2017, 37(2): 244-251.
	[Sun Baisheng, Guo Cuien, Yang Yitianet al. Spatial distribution characteristics of rural place-name cultural landscape based on GIS approach in Chengde. Scientia Geographica Sinica, 2017, 37(2): 244-251.] .
[14]	陶卓霖,程杨,戴特奇,等. 基于公平最大化目标的2020年北京市养老设施布局优化[J]. 地理科学进展, 2015, 34(12): 1609-1616. doi: 10.18306/dlkxjz.2015.12.009
	[Tao Zhuolin, Cheng Yang, Dai Teqi et al. Spatial optimization of residential care facility locations in 2020 in Beijing: Maximum equity in accessibility. Progress in Geography, 2015, 34(12): 1609-1616.] doi: 10.18306/dlkxjz.2015.12.009
[15]	Wang F, Luo W.Assessing spatial and nonspatial factors for healthcare access: Towards an integrated approach to defining health professional shortage areas[J]. Health &Place, 2005, 11(2): 131-146. doi: 10.1016/j.healthplace.2004.02.003 pmid: 15629681
[16]	柳泽, 杨宏宇, 熊维康, 等. 基于改进两步移动搜索法的县域医疗卫生服务空间可达性研究[J]. 地理科学, 2017, 37(5): 728-737.
	[Liu Ze, Yang Hongyu, Xiong Weikang et al.Spatial accessibilities of medical services at county level based on optimized two-step floating catchment area method. Scientia Geographica Sinica, 2017, 37(5): 728-737.]
[17]	戴学珍,徐敏,李杰. 京沪高速铁路对沿线城市效率和空间公平的影响[J]. 经济地理, 2016, 36(3): 72-77. doi: 10.15957/j.cnki.jjdl.2016.03.010
	[Dai Xuezhen, Xu Min, Li Jie.The efficiency and spatial equity impacts of Beijing-Shanghai high-speed rail on neighboring cities. Economic Geography, 2016, 36(3): 72-77] doi: 10.15957/j.cnki.jjdl.2016.03.010
[18]	黎夏,叶嘉安. 遗传算法和GIS结合进行空间优化决策[J]. 地理学报, 2004, 59(5): 745-753. doi: 10.3321/j.issn:0375-5444.2004.05.013
	[Li Xia, Ye Jiaan.Optimal spatial search using genetic algorithms and GIS. Journal of Geographical Science, 2004, 59(5): 745-753.] doi: 10.3321/j.issn:0375-5444.2004.05.013
[19]	李全亮. 免疫算法在带时间窗的车辆路径问题中的应用[J]. 系统工程理论与实践, 2006, 26(10): 119-124. doi: 10.3321/j.issn:1000-6788.2006.10.018
	[Li Quanliang.Immune algorithm for vehicle routing problem with time windows. System Engineering Theory and Practice, 2006,26(10): 119-124.] doi: 10.3321/j.issn:1000-6788.2006.10.018
[20]	张鸿辉,曾永年,尹长林,等. 城市土地利用空间优化配置的多智能体系统与微粒群集成优化算法[J]. 武汉大学学报(信息科学版), 2011, 36(8): 1003-1007.
	[Zhang Honghui, Zeng Yongnian, Yin Changlin et al. Simulating spatial optimization allocation of urban land use based on the integration of multi-agent system and particle swarm optimization algorithm. Geomatics and Information Science of Wuhan University, 2011, 36(8): 1003-1007.]
[21]	张弛,贾丽媛,王加阳. 改进的混合免疫算法在约束函数优化中的应用[J]. 中南大学学报(自然科学版), 2016, 47(6): 1940-1946. doi: 10.11817/j.issn.1672-7207.2016.06.016
	[Zhang Chi, Jia Liyuan, Wang Jiayang.An improved hybrid immune algorithm for multimodal optimization. Journal of Central South University (Science and Technology), 2016, 47(6): 1940-1946.] doi: 10.11817/j.issn.1672-7207.2016.06.016
[22]	上海市城乡建设和交通发展研究院. 2015年上海市综合交通年度报告(摘要)[J]. 交通与运输, 2015,31(6):7-11. doi: 10.3969/j.issn.1671-3400.2015.06.003
	[Shanghai Urban and Rural Construction and Transportation Development Research Institute. Shanghai comprehensive transportation annual report in 2015 (Abstract). Traffic & Transportation,2015,31(6):7-11.] doi: 10.3969/j.issn.1671-3400.2015.06.003
[23]	Peeters D, Thomas I.Distance predicting functions and applied location-allocation models[J]. Journal of Geo-graphical Systems, 2000, 2(2): 167-184. doi: 10.1007/PL00011453

统计指标	β=0.80	β=1	β=1.20	β=1.40	β=1.6	β=1.80	β=2	β=2.20	β=2.40
最大值	2.11	2.92	3.98	5.25	6.66	8.14	9.61	11.03	12.36
最小值	0.64	0.58	0.53	0.44	0.34	0.26	0.19	0.14	0.10
标准差	0.60	0.43	0.65	0.95	1.18	1.44	1.69	1.92	2.13

	现有方案	遗传算法	粒子群算法	模拟退火算法	一般免疫算法	改进免疫算法
政府公平性（张/人）²	295.34	95.98	99.66	113.48	95.44	94.16
配置效率（%）	-	12.80	10.48	10.38	12.67	15.94
居民出行成本（m/人）	660.98	489.21	518.64	540.69	486.23	481.41
投资方收益（元/月）	109200	112500	110204	109200	112500	112500
达到最优值的迭代次数（次）	-	360	280	410	400	220
最优亲和力值	-	1.83	1.89	1.92	1.81	1.78