Utilizing GIS and Remote Sensing Technologies for Site Selection of Rehabilitation and Convalescent Facilities in the North China Plain

https://doi-001.org/1025/17591309085733

JiaHua Liang^1,a, Huan Li^1,b,  Ao Jiao^1,c,  ZhiHai Dong ^2,d,*

¹ Precision Forestry Key Laboratory of Beijing, Beijing Forestry University,

Beijing 100083, China

² Beijing Institute of Surveying and Mapping, Beijing, 100043, China

^aEmail:13111692821@163.com

^bEmail:lihuan7rib@163.com

^cEmail:jiao0318@bjfu.edu.cn

^{* d}Email: 15603285728@163.com

Funding

This study was supported by  5·5 Engineering Research & Innovation Team Project of Beijing Forestry University (BLRC2023A03) and the Natural Science Foundation of Beijing (8232038, 8234065) .

Abstract: In response to the increasingly severe challenge of social aging in the new era, and by integrating the various regulations and systems currently proposed by the state, the construction of an age-friendly living environment has become a focal point of social development. As a special demographic group, the elderly exhibit physiological characteristics, behavioral traits, and lifestyle habits that impose unique demands on their living environment. Therefore, establishing rehabilitation and convalescent zones based on the actual needs of the elderly and utilizing existing ecological and material resources represents one effective approach to addressing social elderly care and medical issues. Amidst rapid advancements in modern technology, this paper integrates GIS and remote sensing techniques to conduct site selection for rehabilitation and convalescent facilities in the North China Plain. It rapidly screens factors influencing residential site selection, followed by performance analysis after model construction.

 Keywords: GIS; remote sensing technology; North China Plain; rehabilitation and convalescence facilities

1. Introduction

Against the backdrop of increasingly severe population aging, the construction and application of rehabilitation and convalescent facilities have garnered significant attention from all sectors of society ^[1]. Among these, integrated rehabilitation community care facilities represent a novel model of elderly care. They provide seniors with diverse services including rehabilitation, long-term care, and community engagement, effectively meeting their daily living and recreational needs. Consequently, they play a crucial role in urban planning and development^([2]).. The specific structure is shown in Figure 1,Key considerations for the construction and application of rehabilitation and convalescent facilities include: first, analyzing geographic distribution trends; second, examining the diversification of client demographics; and finally, understanding energy-saving and environmentally friendly construction requirements. Based on these factors, facility siting and layout must adhere to fundamental principles: facilitating elderly mobility, meeting service needs, aligning with market demands, and preserving environmental and urban aesthetics ^[3].

Figure 1 Comprehensive rehabilitation community old-age care facilities

The North China region refers to the vast area of China north of the Qinling Mountains-Huai River line and south of the Great Wall, bordering Northeast China and the Inner Mongolia Autonomous Region to the north ^[4]. Politically and economically, it encompasses five provincial-level administrative units: Beijing, Tianjin, Hebei, Shanxi, and the Inner Mongolia Autonomous Region, constituting the largest and most dynamic economic zone in northern China ^[5]. Based on this, selecting locations for rehabilitation-oriented community integrated elderly care facilities involves two layout models: vertical and horizontal. Each has distinct advantages and disadvantages in application. Specific considerations must integrate rational conditions, facility characteristics, and service demands to comprehensively explore how best to meet the needs of the elderly population, market development requirements, and urban ecological environments. With the continuous advancement of modern technology, scholars have begun integrating Geographic Information Systems (GIS) and remote sensing techniques to develop more rational site selection strategies for planning and constructing rehabilitation-oriented community-integrated elderly care facilities ^[6].

2. Research Methodology

2.1 Evaluation Method

Selecting sites for rehabilitation facilities in the North China Plain requires the organic integration of GIS with multi-criteria decision-making methods, typically combining GIS with fuzzy analytic hierarchy process (AHP). To ensure the model better aligns with practical needs, a fuzzy decision-making trial and evaluation experimental model is introduced. This model constructs relationships between various indicators and optimizes their weight values, ultimately yielding comprehensive indicator weights ^[7]. Referring to the model calculation flowchart shown in Figure 1, the practical operational steps are as follows:

Figure 1: Model Evaluation Flowchart

Following this process, first standardize all indicator layers within the environment. Normalize layer data using methods like Gaussian fuzzy membership functions to ensure values range from 0 to 1. Subsequently, employ AHP and the Fuzzy Trial and Evaluation Method to derive the composite weights for each indicator layer. Finally, perform weighted processing on the nine layers to obtain the final composite layer ^[8].

2.2 Analytic Hierarchy Process (AHP)

As a systematic and structured process, the Analytic Hierarchy Process (AHP) primarily involves the following steps, as illustrated in Figure 2:

Figure 2: Flowchart of the Analytic Hierarchy Process

First, construct a hierarchical structure model. Decision-makers must thoroughly analyze the problem to precisely identify its key factors, thereby establishing a clear hierarchical structure. This logical framework reveals the problem’s essence and provides a solid foundation for subsequent computational analysis ^[9].

Second, construct the judgment matrix. After establishing the hierarchical structure, pairwise comparisons are performed among elements within the same level to determine their relative importance to a criterion in the preceding level ^([10]). For n elements C₁, C₂,…, C_nat a given level, these comparisons yield an n×n square matrix—the judgment matrix A—as illustrated below:

In this model, a_ijrepresents the importance scale value of element C_irelative to element C_j. This matrix must satisfy two conditions to form a positive-inverse matrix: first, diagonal elements equal 1 ( ); second, elements are reciprocals of each other ( ).

Third, compute the weight vector. Based on the aforementioned judgment matrix A, the relative weight vector W = (w₁, w_(2),…, w_n)^Tcan be derived for each element. This constitutes the core calculation of the Analytic Hierarchy Process ^[11].

Theoretically, the weight vector W of the judgment matrix A corresponds to the eigenvector associated with its largest eigenvalue. By analyzing the eigenvalue equation, the specific eigenvector equation is obtained as follows:

Based on this, normalizing vector W to ensure the sum of all its components equals 1 yields the weights for each element. Theoretically, this method is the most precise and is known as the Eigenvector Method ^[12].

Scholars have proposed approximate calculation methods, primarily applied when dealing with high-order matrices or lacking specialized software. While precisely solving for eigenvectors involves complex steps, approximate methods ensure consistent results that closely approximate those of the eigenvector method ^[13].

The geometric mean method is calculated as follows:

In arithmetic mean analysis, since each column of the judgment matrix A approximates the distribution of weight values, the arithmetic mean of all column vectors can be used to estimate the analysis vector. The specific formula is as follows:

Since judgment matrices are based on subjective input and may contain logical inconsistencies, the consistency of judgment evidence must be rigorously examined. The practical steps are as follows: First, calculate the maximum eigenvalue. For a perfectly consistent n×n positive definite matrix, the maximum eigenvalue is ; for inconsistent matrices, it is:

Second, when calculating the Consistency Index (CI), a higher value indicates greater inconsistency in the judgment matrix. The specific formula is as follows:

Third, determine the average random consistency index. This value is obtained by averaging the consistency indices of a large number of randomly generated positive definite matrices. The actual result is closely related to the order of the matrix ^[14].

Fourth, calculate the Consistency Ratio (CR) using the following formula:

Fifth, consistency judgment. If the consistency ratio is less than 0.10, the consistency of the judgment matrix is acceptable. Otherwise, it indicates severe logical contradictions in the judgment matrix, requiring decision-makers to re-examine and revise the negative values of pairwise comparisons until the consistency ratio meets the requirement. After completing weight calculations and consistency checks across all levels, the final step involves determining the ultimate weights of each alternative at the scheme level relative to the overall objective. This process is termed hierarchical total ranking ^[15].

Assuming the weight vector of criterion layer C relative to objective layer O is:

The weight vector of the P-th scheme layer P relative to the k-th criterion C_kis:

Then the total weight of the i-th proposal P_iin the proposal layer is:

2.3 Screening Factors

After comprehensively considering the site selection requirements for rehabilitation and elderly care facilities, CiteSpace identified four positive utility factors: first, a higher floor area ratio; second, a higher population density; third, distance from noise disturbances; and finally, distance from air pollution. Simultaneously, it identified one primary factor yielding negative benefits for the elderly population: housing prices ^[16].. Since the elderly primarily rely on walking and public transportation for mobility, the travel costs examined in this study primarily involve the spatial distance between the selected location and facilities closely related to the elderly population. The distance between these facilities and senior residences affects the ease of travel for the elderly, directly impacting their overall life satisfaction. Ideally, the location of rehabilitation and convalescent facilities should maximize fulfillment of the corresponding needs of the elderly population. Therefore, through literature review, facilities closely related to the elderly and not easily distributed in large quantities are categorized as follows: First, healthcare facilities, such as those providing preventive, medical, health, and rehabilitation services directly to the community, primarily addressing the daily needs of the elderly for common illnesses, chronic diseases, and frequently occurring conditions; Second, recreational and leisure facilities, such as urban ecosystems and landscapes, which provide leisure spaces for the elderly, along with outdoor sports venues that facilitate physical exercise. Finally, commercial and daily service facilities, including restaurants, home-based elderly care service centers, and cultural exhibition halls ^[17].

2.4 Model Construction

Due to the differing dimensions and independence of utility indicators, it is impossible to directly form a unified metric to measure the positive utility of resident travel. Therefore, when constructing the model, dimensionless calculation methods must be employed to eliminate influencing factors. The specific formula is as follows:

In the above formula, I’ denotes the dimensionless value of indicator I, and I_maxrepresents the maximum value within indicator I. The positive utility calculation formula is as follows:

In the above formula, represents the dimensionless value of utility 1 for location j. Assuming utility 1 in this study represents floor area ratio (FAR), and denotes the weighting proportion of FAR among the five indicators, the utility value _P(j)for location _j can be obtained by summing the five utility indicators using the above method, where housing prices are taken as negative values ^[18].

The cost studied in this paper refers to the cost in the generalized domain, i.e., the total expenditure of elderly patients during travel. The specific formula is as follows:

In the above formula, C_ijrepresents the travel cost between points i and j, where F_ij, T_ij, and K_ndenote the travel expenses, travel time, and comfort level between points i and j, respectively; VOT indicates the time value of travel for the group. As the study focuses on the elderly population, only time costs are analyzed.  denotes the dimensionless value of travel costs between points i and j, while C_maxrepresents the maximum generalized travel cost among all point pairs.

The utility function is selected as follows:

In the above formula, the parameter represents the influence level of the generalized cost. Its value range varies across different studies. Based on existing scholarly research, this study sets its value range between 1.5 and 2.0, where it has the least impact on the research results.

3. Experimental Results

3.1 Weighting System

The impact of various utility and cost factors on rehabilitation and convalescent facilities for the elderly varies significantly. This study employs the Analytic Hierarchy Process (AHP) to generate a weighting system, identifying factors with significant influence on the elderly. These weights serve as essential support for constructing the utility model, yielding the analytical results presented in Tables 1, 2, and 3 below:

Table 1: Analytic Hierarchy Process Judgment Matrix for Utility Factors

Table 2: AHP Results for Utility Factors

Table 3 Consistency Test Results

Analysis of the above tables reveals that the most significant utility factor for selecting rehabilitation and convalescent facility locations is distance from noise pollution, followed by floor area ratio. Population density and housing prices carry relatively lower weights. Additionally, healthcare and daily living service facilities exert highly significant influence on site selection.

3.2 Final Site Selection

This study downloaded maps of the research area from OpenStreetMap as a primary reference. Subsequently, information on various facilities—including noise/exhaust pollution control facilities, healthcare facilities, commercial facilities, and recreational/entertainment facilities—was obtained using AutoNavi Maps. Population density data was downloaded from the Resource and Environmental Science Data Registration and Publication System, while key information such as housing prices and floor area ratios across different regions was also gathered. ArcGIS was then employed to conduct an in-depth analysis of existing facilities and population distribution in the North China Plain region ^([19]).The weighting system derived from the table above indicates that selecting three relatively stable factors yields a more consistent utility distribution for site selection. The negative impact of rail transit is particularly pronounced, resulting in higher utility values for both city centers and remote suburban areas. This demonstrates that these locations offer greater fixed benefits for the daily lives of the elderly population, possessing inherent potential for establishing rehabilitation and convalescent facilities. Simultaneously, integrating the relatively variable factors of housing prices and floor area ratio enables more precise site selection for geriatric rehabilitation and convalescent facilities. The final results indicate that highly suitable areas possess all key factors identified in the table above. In other words, these locations feature high facility density with diverse types, low commuting costs, sufficient distance from pollution sources, city walls that effectively block pollution dispersion, relatively high population density enhancing facility stability, and moderate surrounding housing prices that avoid significant negative impacts ^[20] . From a long-term development perspective, this area represents the most suitable foundation for locating rehabilitation and convalescent facilities on the North China Plain at present.

 Conclusion

In summary, this study systematically outlines the site selection process for rehabilitation and convalescent facilities—covering methodology screening, factor selection, landscape modeling, and site confirmation—using the North China Plain as a case study. It provides a reference framework for social development amid the increasingly severe challenges of population aging in the new era.

 References

1. Hongyu Zhang, Zoufeng Li, Ying Liu. A GIS-Based Methodology for Selecting Interchange Locations Between Expressways and Urban Roads [J]. Hunan Journal of Transportation Science and Technology, 2023, 49(4):192-196.
2. Xiaofang Lu, Yishan Zuo, Jianfeng Liu. GIS-Based Study on the Site Selection and Layout of Off-Street Parking Facilities in Old Urban Areas: A Case Study of Gongyi City’s Old District [J]. Urban Architecture, 2023, 20(18):146-148.
3. Ting Huang, Yaozong Lu, Lejun Yu. Evaluation of Potential for Utilizing Shenzhen’s Spatial Resources as Shelters Based on GIS-MCDA [J]. Surveying and Spatial Information, 2024, 47(11):111-114.
4. Weimin Huang, Jun Zheng, Ke Wu, et al. Development and Application of a GIS-Based Decision Support System for Shelter Planning [J]. Science and Technology Bulletin, 2023, 39(12):85-89. DOI:10.13774/j.cnki.kjtb.2023.12.016.
5. Yuxiang Han, Liang Chen. Application of GIS-Based Kernel Density Analysis in Offshore Buoy Platform Site Selection [J]. China Maritime Science, 2023, 46(3):59-64.
6. Yu Zhong, Yingfu Li, Chao Lei, et al. GIS-Based Site Selection for Regional Cigarette Logistics Distribution Centers [J]. Tobacco Science and Technology, 2024, 57(4):91-99.
7. Yongshi Hu, Shuhao Lü, Chenyang Qin. Site Selection for Trade-Hub-Type Aquatic Product Logistics Centers in Coastal Regions [J]. Transportation Science and Economy, 2024, 26(1):34-42.
8. Peng Xie, Yunfeng Huang, Pengcheng Yu. GIS-Based Study on Spatial Accessibility of Medical Services in Hefei’s Central Urban Area [J]. Journal of Henan University of Urban Construction, 2024, 33(3):65-71.
9. Yong Gao, Weicheng Zheng, Zhengbin Wei. Research and Application of Macro-Level Wind Farm Site Selection Method Based on GIS Analysis [J]. China New Technology and New Products, 2023(7):125-128. DOI:10.3969/j.issn.1673-9957.2023.07.041.
10. Ji Wang, Zheng Niu, Ni Huang, et al. Optimization of Medical Institution Site Selection Combining GIS and Analytic Hierarchy Process [J]. Science and Technology Innovation, 2023(3):201-206.
11. Jian Wang. Application of GIS Technology in Airport Site Selection Research [J]. Intelligent City Applications, 2024, 7(12):119-122. DOI:10.33142/sca.v7i12.14728.
12. Shiqi Liu, Hang Mo, Shuxun Sang, et al. Source-Sink Matching and Cluster Deployment for Carbon Capture, Utilization, and Storage in Ningxia Hui Autonomous Region [J]. Journal of Coal Science and Engineering, 2024, 49(3):1583-1596. DOI:10.13225/j.cnki.jccs.XH23.1378.
13. Zhixia Zhang, Pengzhang Li. Application of Improved HHO Algorithm in Location Selection for Perishable Goods Distribution Centers [J]. Complex Systems and Complexity Science, 2024, 21(4):91-98. DOI:10.13306/j.1672-3813.2024.04.014.
14. Xiaohui Huang. Hospital Site Selection Decision and Optimization Model Based on AHP and GIS: A Case Study in a District of Ganzhou City [J]. Heilongjiang Science, 2024, 15(4):70-72. DOI:10.3969/j.issn.1674-8646.2024.04.017.
15. Wenyuan Yu. Design of a GIS-Based Public Parking Lot Location Model: A Case Study in Licheng District, Jinan City [J]. Design (Hans), 2023, 8(4):4000-4008.
16. Rong Wu, Guoxian He. Simulation Study on Dynamic Site Selection for Material Logistics Distribution Centers in Construction Supply Chains [J]. Construction Economics, 2023, 44(S02):583-587.
17. Haoran Lin, Weijun Xu, Zhen Zhang. Spatial Evolution Analysis of Site Selection Behavior for “Specialized, Refined, Distinctive, and Innovative” Small Giant Enterprises: A Case Study of Guangdong Province [J]. New Economy, 2024(9):87-100.
18. Shu Yang, Zhenhua Yang, Mingyu Tian, et al. A Brief Discussion on the Application of GIS Technology in the Site Selection of Emergency Shelters [J]. Science and Technology Innovation, 2025(1):84-87.
19. Jinghua Dai, Yuji Qin, Yang Qu, et al. Application Analysis of GIS System Based on Utility Theory in Senior Housing Site Selection [J]. Urban Architecture, 2024, 21(21):106-111.
20. Mingqi Guo, Hailing Li. GIS-Based Multi-Source Data Planning for Affordable Rental Housing Site Selection: A Case Study of Sichuan Tianfu New Area [J]. Construction Engineering (Chinese and English Edition), 2024, 8(2):1-10.