Cities 144 (2024) 104637

Contents lists available at ScienceDirect

Cities
journal homepage: [Link]/locate/cities

Supply-demand matching assessment of the public service facilities in

15-minute community life circle based on residents' behaviors
Lu Song a, Xuesong Kong a, b, *, Peng Cheng c
a
School of Resource and Environmental Sciences, Wuhan University, Wuhan 430079, China
b
Hubei Luojia Laboratory, Wuhan 430079, China
c
School of Management, Wuhan Institute of Technology, Wuhan 430205, China

A R T I C L E I N F O A B S T R A C T

Keywords: 15-minute community life circle focuses on residents' daily needs and aligns with “people-oriented” planning.
Community life circle However, the personalized needs of residents have been ignored in literatures, resulting in the mismatch between
Public service facility the supply and demand of public service facilities (PSFs) as well as the unfair supply of different population
Behavioral demand
groups among communities. Based on community life circle, we proposed a multi-group huff three-step floating
Two-step floating catchment area
catchment area (MGH3SFCA) model to evaluate the supply-demand matching levels of PSFs for different age
groups in Nanjing, China. The results showed that the supply-demand between PSFs and different age groups was
relatively mismatched in the main urban area of Nanjing, and medical care facilities were seriously inadequate to
fulfill the needs of any age group. Comprehensive accessibility varied depending on the frequency of visits to
PSFs for each age group. Although age groups had a similar spatial pattern of comprehensive accessibility at
community and building scales, the service coverage rate of PSFs was larger at the building scale, and the
assessment of supply-demand matching of PSFs at the building scale was more accurate and reliable. The findings
would provide important decision-making support for the targeted allocation of PSFs in 15-minute community
life circle.

1. Introduction caters for the actual needs of the post-pandemic era (Allam, Bibri,
Chabaud, & Moreno, 2022; Allam, Bibri, Jones, et al., 2022). Up to now,
Public service facilities (PSFs), such as educational and medical fa­ the concept has been put into practice in many cities worldwide,
cilities, provide various public goods and services to fulfill residents' including Paris, Tempe and Melbourne (Logan et al., 2022; Oostenbach
daily needs (Cui et al., 2022). Rational allocation of PSFs is of great et al., 2023).
significance for improving residents' quality of life and achieving social As China enters the middle and late stage of urbanization (Cheng
equity (Han et al., 2023; Xia et al., 2022). Sustainable Development Goal et al., 2023), the process of urbanization has shifted from high-speed
(SDG) 11.7 aims to provide universal access to safe, inclusive and growth to high-quality development (Liu & Zhu, 2022), and urban
accessible, green and public spaces, in particular for women and chil­ planning has gradually shifted from “place-oriented” to “people-ori­
dren, older persons, and persons with disabilities (United Nations, ented” (Li et al., 2021). Improving urban living environment and resi­
2015). However, the existing allocation of PSFs ignores the personalized dents' quality of life have become new requirements for urban
needs of residents, resulting in the mismatch between the supply and development (Luo et al., 2022). In this context, community life circle
demand of PSFs and the unfair supply of different population groups planning, which focuses on residents' daily activity space, has become an
among communities (Dadashpoor et al., 2016; Weng et al., 2019; Zheng important way to promote the equalization of basic public services and
et al., 2020). The COVID-19 pandemic has brought these problems to the achieve sustainable development of urban society (Liu & Chai, 2015). In
fore, and innovative measures are urgently needed to ensure that resi­ 2016, Shanghai, China, took the lead in putting forward the concept of
dents have equal access to the most basic PSFs (Burke et al., 2022). 15-minute community life circle (Shanghai Planning and Land Re­
Community life circle proactively responds to the differentiated needs of sources Administration, 2016). Subsequently, the standard for urban
residents, which aligns with “people-oriented” planning (Yu, 2019) and residential area planning and design (Ministry of Housing and Urban-

* Corresponding author at: School of Resource and Environmental Sciences, Wuhan University, 129 Luoyu Road, Wuhan 430079, China.
E-mail address: xuesongk@[Link] (X. Kong).

[Link]
Received 21 February 2023; Received in revised form 2 October 2023; Accepted 17 October 2023
Available online 29 October 2023
0264-2751/© 2023 Elsevier Ltd. All rights reserved.
L. Song et al. Cities 144 (2024) 104637

Rural Development of the People's Republic of China, 2018) and the but few studies evaluate the accessibility to PSFs at the building scale
spatial planning guidance to community life unit (Ministry of Natural due to the lack of fine-grained population data (Jing et al., 2023; Zhong
Resources of the People's Republic of China, 2021) have been released et al., 2020). Second, the spatiotemporal behaviors of different popu­
by the Chinese government to guide the allocation of PSFs. Some major lation groups are ignored in community life circle, which cannot effec­
cities in China, such as Beijing and Wuhan, have proposed to build 15- tively respond to the differentiated needs of residents (Khavarian-
minute community life circles to meet the daily needs of residents Garmsir et al., 2023; Zou et al., 2021).
(Liu, Zheng, et al., 2022; Luo et al., 2022). To fill the abovementioned research gap, this paper aims to analyze
Community life circle is a space range carrying out various daily the behavioral demand characteristics of different age groups and
activities centered around residents' homes (Xiao et al., 2014). In 1941, evaluate their supply-demand matching levels of PSFs in 15-minute
Ishikawa came up with the “life circle composition” based on the central community life circle. We proposed a multi-group huff three-step
place theory (Sun & Chai, 2018). In 2016, Moreno et al. (2021) proposed floating catchment area (MGH3SFCA) model to quantify the accessi­
the “15-minute City”, which gained widespread attention again during bility to PSFs for different age groups at community and building scales
the COVID-19 pandemic (Khavarian-Garmsir et al., 2023). Studies of in Nanjing, China. The main objectives are to: (i) quantify the accessi­
community life circle mainly focus on two aspects: scope delineation and bility to PSFs for different age groups; (ii) analyze the difference in
facilities allocation. accessibility among different age groups; and (iii) propose imple­
In terms of scope delineation, traditional studies usually delineate mentation paths for 15-minute community life circle planning.
the scope of community life circle according to the existing adminis­
trative boundaries and indicators such as walking time, population size, 2. Materials and methods
and land area (Guo et al., 2019), which cannot cater for the personalized
needs of residents. Based on Global Positioning System (GPS) data, the Fig. 1 presents the overall methodological framework. Firstly, the
Alpha-shape method (Sun et al., 2016) and “context-based crystal- kernel density estimation (KDE) was used to analyze the spatial distri­
growth activity space” method (Chai et al., 2019) are applied in delin­ bution characteristics of various PSFs. Secondly, according to the age
eating the range of community life circle by identifying residents' ac­ structure, residents were divided into four groups (i.e., infants, children,
tivity space and analyzing residents' behavioral demands. However, GPS adults, and seniors) to analyze the spatial distribution and behavioral
data is only suitable for small-range delineation due to the high cost of demand characteristics of different age groups. Finally, the MGH3SFCA
data acquisition. It is urgent to explore a community life circle delin­ was proposed to calculate the accessibility to various PSFs for different
eation method which not only caters for the concept of “people-ori­ age groups, and then the entropy method was applied to calculate the
ented”, but also can be widely applied in practice (Li et al., 2021). With comprehensive accessibility to analyze the difference among different
the rapid development of mobile phone signaling data, several scholars age groups.
have begun to use it to investigate the scope of residents' daily life (Chen,
Li, et al., 2022; Lin et al., 2021; Xiao et al., 2019; Zhao et al., 2021,
2022). Mobile phone signaling data, which is large sample size, fast data 2.1. Study area and data description
update and precise in terms of spatial location (Liu, Meng, et al., 2022),
has full potential to support community-scale research (Zhang et al., Nanjing, the capital of Jiangsu Province in eastern China, is one of
2023). the few aging cities with a century-old planning history. In 2020,
In terms of facilities allocation, a variety of methods have been Nanjing had a total population of 9.31 million, of whom the urban
developed to measure the accessibility to PSFs from the perspective of population accounted for 86.80 %, the population aged 0–14 accounted
equity and efficiency, including network analysis (Kesarovski & for 12.75 %, the population aged 15–59 accounted for 68.27 %, and the
Hernández-Palacio, 2022), walk score (Weng et al., 2019), cumulative population aged 60 and above accounted for 18.98 %. The aging pop­
opportunity measure (Calafiore et al., 2022), gravity model (Xia et al., ulation will keep increasing, which poses a huge challenge to urban
2018), two-step floating catchment area (2SFCA) method (Hu, Zhao, planning. In 2021, Nanjing was selected as one of the first national pilot
et al., 2020; Hu, Song, et al., 2020), optimal supply-demand allocation cities for building 15-minute community life circles. It is of great sig­
model (Song et al., 2022), etc. For example, Weng et al. (2019) utilized nificance for Nanjing to provide equal access to PSFs, in particular for
the walk score metric to measure the walkability of 15-minute neigh­ vulnerable groups, such as children and seniors. The main urban area of
borhoods. Hu, Zhao, et al. (2020) utilized the enhanced 2SFCA Nanjing, as the key region to improve residents' quality of life, was
(E2SFCA) method to calculate the dynamic healthy food accessibility. selected as the study area (Fig. 2). The scope of the main urban area
The 2SFCA method incorporates the supply of facilities, the demand of refers to the Overall Planning of Nanjing (2011− 2020), including
residents, and the distance between supply and demand into the esti­ Gulou, most areas of Qinhuai, Xuanwu and Jianye, and a small part of
mation of accessibility, which can unravel the difference in spatial dis­ Yuhuatai, Qixia and Jiangning District. The southeast of Gulou, south­
tribution at the micro scale and reflect the balance between supply and west of Xuanwu and northwest of Qinhuai District belong to the Old
demand at the macro scale (Wu et al., 2020). Since the initial 2SFCA Town, which is the historic, cultural and commercial center of Nanjing
method is greatly affected by the threshold travel time or distance (Luo (Chen, Chen, et al., 2022).
& Wang, 2003), various extensions have been proposed in subsequent The data used in our study mainly included communities, road net­
studies (Tao & Cheng, 2016), including introducing distance decay works, population, buildings and points of interest (POIs) in the main
function (Dai, 2010, 2011; Luo & Qi, 2009), rationally setting service urban area of Nanjing.
radius (Dony et al., 2015; Luo & Whippo, 2012; McGrail & Humphreys,
2014), incorporating the potential for competition among facilities (Luo, (i) POIs data: This data was collected by Nanjing Municipal Bureau
2014; Wan et al., 2012), and considering multiple transportation modes of Planning and Natural Resources in 2020. By referring to the
(Langford et al., 2016; Mao & Nekorchuk, 2013). Recently, scholars current planning standards of community life circle (Ministry of
have integrated different types of improvements within the family of Housing and Urban-Rural Development of the People's Republic
2SFCA methods into one method (Jörg & Haldimann, 2023; Subal et al., of China, 2018; Ministry of Natural Resources of the People's
2021; Wang et al., 2021). For example, supply competition and absolute Republic of China, 2021; Nanjing Municipal Bureau of Planning
differences in distances were incorporated into the modified huff three- and Natural Resources, 2023; Nanjing Municipal People's Gov­
step floating catchment area (MH3SFCA) method (Subal et al., 2021). ernment, 2023), we selected 6 categories and 11 subcategories of
Significantly, current studies have two limitations. First, cities are PSFs closely related to residents' daily life, including education,
characterized by high spatial heterogeneity, even within a community, medical care, elderly care, culture, sports and shopping, and

2
L. Song et al. Cities 144 (2024) 104637

Fig. 1. Analytical framework of the study.

constructed a configuration system based on the three-level infants (age 0–2), children (age 3–17), adults (age 18–59) and
community life circle (Table 1). seniors (age 60+). Furthermore, the group of children was sub­
(ii) Population and building data: Both population and building data divided into preschool children (age 3–5), pupils (age 6–11), ju­
were collected by Nanjing Municipal Public Security Bureau. The nior high school students (age 12–14), and senior high school
population data is vector point data collected in 2021, with at­ students (age 15–17).
tributes of gender and age, while the building data is vector (iii) Road network data: This data of 2021 was downloaded from
polygon data collected in 2020, with attributes of layers and OpenStreetMap ([Link] After
height (Fig. S1). To obtain the demographic data of each building, excluding motorways and other roads unsuitable for walking, we
we used the spatial connection tool in ArcGIS 10.2 to assign corrected topology errors and constructed road networks in
population attributes to the nearest building. The target features ArcGIS 10.2. We used the network analysis tool to establish OD
were set as populations, the join features were set as buildings, cost matrix for each age group. The origin of OD cost matrix was
and the match option was set as closest. If there were multiple set as buildings, and the destination was set as PSFs, so that the
buildings with equal distances from the population, one of the road network distance between each building and facility was
buildings would be randomly selected as the matching feature. calculated.
Considering that the age structure is the main factor affecting
differences in resident's behavioral demands (Li, 2017), residents
were divided into four groups according to the age structure:

3
L. Song et al. Cities 144 (2024) 104637

Fig. 2. The main urban area of Nanjing.

2.2. Kernel density estimation

Table 1
The frequency of visit to various PSFs by different age groups.
The KDE can measure spatial distribution characteristics such as the
Category Subcategory Community Target groups Visit position and density of points by constructing a continuous probability
life circle frequency
density function based on discrete sample points, which is an important
Education Kindergarten 5 min Preschool High tool for studying uneven spatial distribution (Liu, Xue, et al., 2022; Liu &
children Zhu, 2022). In our study, the KDE was used to analyze the spatial dis­
Primary school 10 min Pupils High
Junior high 15 min Junior high High
tribution characteristics of various PSFs and age groups. The kernel
school school density function is estimated by Eq. (1) (Dong et al., 2022).
students ( )
Medical Community 5 min Infants Middle 1 ∑n di
f (x, y) = 2 k (1)
care healthcare clinic Children Low nh i=1 h
Adults Low
Seniors Middle where f(x, y) is the density function of point (x, y); di is the distance
Community 15 min Infants Middle
between point (x, y) and the i th observation point; n is the number of
healthcare center Children Low
Adults Low observation points; k is the kernel function; h is the bandwidth. Ac­
Seniors High cording to the spatial scale of community life circle, the service radius of
Elderly Home-based 10 min Seniors High all PSFs and age groups was set as 1, 500 m.
care service center
Nursing home 15 min Seniors High
Culture Community 15 min Children Middle 2.3. Classification of residents' behaviors based on age groups
cultural center Adults Middle
Seniors Middle According to the age structure, we divided the residents into four
Sports Community 15 min Children Middle
sports center Adults High
groups, and then analyzed the behavioral demand characteristics (i.e.,
Seniors Middle travel distance and visit frequency) of different age groups.
Shopping Fresh market 10 min Adults Middle
Seniors High (i) Travel distance: Community life circle highlights the walking
Supermarket 15 min Children Low
accessibility to PSFs (Liu, Zheng, et al., 2022). Liu et al. (2020)
Adults High
Seniors High believed that 5-minute (400–500 m), 10-minute (800 m–1000 m)
and 15-minute (1, 500–1, 600 m) community life circle would be
applicable in the future. Notably, the slower walking speed of
infants and seniors should be taken into account. For example,
Zhang et al. (2019) set the walking speed of seniors as 3.5 km/h.

4
L. Song et al. Cities 144 (2024) 104637

The Institute for Transportation and Development Policy (2022) ( )

∑ ∑ Sj f dij
reported that when caregivers traveled with infants, they walked Ai = Rj = ∑ ( ) (2)
slower, with an average walking speed of 15 to 60 m/min. In this j∈{dij ≤d0 } j∈{dij ≤d0 }
Pk f dkj
context, for infants and seniors, we set 5-minute, 10-minute and
k∈{dkj ≤d0 }

15-minute walking distance as 300 m, 600 m, and 900 m,

where Rj is the supply-to-demand ratio of facility j; Sj is the capacity of
respectively, and for children and adults, 500 m, 1, 000 m, and 1,
500 m, respectively. facility j; Pk is the population at location k; dkj is the distance between k
( )
(ii) Visit frequency: We determined the frequency of visit to various and j; d0 is threshold travel distance; f dkj is the distance decay func­
PSFs by different age groups (Table 1) based on questionnaires tion.
from the Chinese government and scholars (Feng & Yang, 2015; The generalized 2SFCA method assumes that all residents have the
Li, 2017; Weng et al., 2019). It is worth noting that seniors have a same walking speed and visit PSFs with the same frequency, ignoring the
high participation in collective activities in China (Feng & Yang, differentiated needs of different age groups. Inspired by the MH3SFCA
2015), and they visit PSFs more frequently than other age groups. (Subal et al., 2021) and multi-mode 2SFCA (MM2SFCA) (Langford et al.,
2016; Mao & Nekorchuk, 2013; Zhang et al., 2019), we proposed the
2.4. Calculation of the accessibility to PSFs for different age groups MGH3SFCA method to quantify the accessibility to PSFs for different age
groups {M1 , M2 , …, Mn }. The method incorporates the strength of pre­
The generalized 2SFCA method for evaluating the accessibility to vious measures and consists of three steps (Fig. 3).
PSFs at location i, Ai , is formulated in Eq. (2) (Zhang et al., 2019).
Step 1: Calculate the probability to visit facilities for different age
groups.

Fig. 3. A sketch map of MGH3SFCA for different age groups.

5
L. Song et al. Cities 144 (2024) 104637

Previous studies (Delamater, 2013; Wan et al., 2012) have shown we compared the differences between the MH3SFCA (single-group) and
that ignoring competition among facilities may overestimate the popu­ MGH3SFCA (multi-group) by taking community healthcare centers with
lation demand. So the huff model was introduced to calculate the a large difference in visit frequency among different age groups as an
probability of residents' selection on facilities (Luo, 2014). We divided example.
the population Pk at location k into n subpopulations according to the For MH3SFCA (single-group), we assumed that everyone had the
{ }
age structure, which could be expressed as Pk = Pk,M1 , Pk,M2 , …, Pk,Mn . same needs as adults, and the coefficient reflecting visit frequency β was
For subpopulation Mn at location i, the probability to visit facility j, also introduced for calculation. The maximum walking distance d0 was
Huff ij,Mn , can be calculated by Eq. (3). set as 1, 500 m, and the coefficient β was set as 0.2.
( ) ( )
Sj G dij , d0
Huff ij,Mn = ∑
Sj G dij (Mn ) , d0 (Mn )
(3) Huff ij = ∑ (7)
Sk G(dik (Mn ) , d0 (Mn ) ) Sk G(dik , d0 )
k∈{dik (Mn )≤d0 (Mn ) }
k∈{dik (Mn )≤d0 (Mn ) }

( ) ( )2 ()
where Sj is the capacity of facility j, due to the lack of capacity infor­ ⎧
− 1 ×
dij
− 1
⎪
mation in POIs data, we assumed that the capacity of each facility
2 d0 2
⎪
⎪
( ) ⎨e − e
() , if dij (Mn ) ≤ d0 (Mn )
equaled 1; dij (Mn ) is the road network distance between subpopulation G dij , d0 = (8)
⎪
Mn at location i and facility j; d0 (Mn ) is the maximum walking distance
− 1
⎪
⎪ 2
( ) ⎩ 1− e
for subpopulation Mn (i.e., defined in Section 2.3); G dij (Mn ) , d0 (Mn ) is
0, if dij (Mn ) > d0 (Mn )
Gaussian function, which can be calculated by Eq. (4).
() ( )2 () 1000Sj
⎧ 1 × dij (Mn ) 1
Rj = ∑ (9)
⎪
⎪
− 2 d0 (Mn )
− 2 Huff ik βPi
⎪e
⎨ − e i∈{dik ≤d0 }
( ) () , if dij (Mn ) ≤ d0 (Mn )
G dij (Mn ) , d0 (Mn ) = ∑
⎪ 1 ( )
(10)
⎪ −
⎪
⎩ 1− e
2
Ai = Huff ij G dij , d0 Rj
j∈{dij ≤d0 }
0, if dij (Mn ) > d0 (Mn )
(4) For MGH3SFCA (multi-group), the accessibility for different age
groups Ai,Mn was weighted by the size of subpopulation and the coeffi­
Step 2: Calculate the supply-to-demand ratio for each facility. cient reflecting visit frequency to calculate the composite accessibility at
position i.
For each facility j, the supply-to-demand ratio Rj can be calculated by ∑Mn
M Ai,Mn β(Mn )Pi,Mn
Eq. (5). As assuming an optimal configuration leads to an overestimation Ai = ∑1 Mn (11)
M1 β(Mn )Pi,Mn
of the overall accessibility (Delamater, 2013), similar to the MH3SFCA,
we removed the distance weight in the denominator to consider the
absolute differences in distances. Additionally, we quantified the fre­ 3. Results
quency of visit to various PSFs by different age groups to cater for res­
idents' actual needs. 3.1. Spatial distribution of PSFs and age groups

Rj = ∑Mn ∑
1000Sj
(5) In general, the spatial distribution of most PSFs was relatively
M1 Huff ik,Mn β(Mn )Pi,Mn balanced, with high-density areas mainly concentrated in the Old Town
(Fig. 4). Among the six types of PSFs, the number of medical care fa­
i∈{dik (Mn )≤d0 (Mn ) }

cilities was the lowest, and the distribution was the most inequitable.
where Pi,Mn is the size of subpopulation Mn at location i; β(Mn ) is a co­
Notably, community healthcare clinics, as essential facilities in 5-minute
efficient reflecting visit frequency. We divided visit frequency into low,
community life circle, should be supplied adequately in space. However,
medium and high, and set β(Mn ) as 0.2, 0.5 and 0.8, respectively. Due to
there were only 40 community healthcare clinics, which was less than
the low value of “per person”, we used “per thousand” instead of “per
the number of community healthcare centers (i.e., 51), and they were
person” via multiplying the numerator by 1, 000.
distributed unevenly among districts.
The spatial distribution of different age groups affects their accessi­
Step 3: Calculate the accessibility for different age groups.
bility to PSFs (Luo & Wang, 2003). Fig. 5 showed that the population
density was high in the Old Town, and it had to cope with a severe aging
Based on the supply-to-demand ratio Rj , for subpopulation Mn at
issue (Zhang et al., 2019). Additionally, by comparing the four age
location i, the accessibility Ai,Mn can be calculated by Eq. (6).
groups, an interesting phenomenon could be found in the central part of
∑ ( )
Ai,Mn = Huff ij,Mn G dij (Mn ) , d0 (Mn ) Rj (6) Jianye District, where the population density of adults was very high,
j∈{dij (Mn )≤d0 (Mn ) } infants and children were also clustered. The central part of Jianye
District is Nanjing Hexi Central Business District, which is the core area
Since there is usually no cross-district school enrollment in reality, it for Nanjing to build a financial center in the Pan-Yangtze River Delta,
is necessary to calculate the accessibility to educational facilities based with a focus on developing modern service industries such as finance,
on each district (and if possible, each school district). Besides, we culture and sports. It is predicted that more and more adults will settle
defined the supply-to-demand ratio Rj of educational facilities as fol­ here, the number of infants and children is also expected to increase.
lows: (i) Rkindergarten = the number of kindergartens / the number of However, combined with Figs. 4 and 5, it could be found that the supply-
preschool children; (ii) Rprimary school = the number of primary schools / demand between some PSFs and age groups did not match. Neither PSFs
the number of pupils; (iii) Rjunior high school = the number of junior high that adults often visit, such as community sports centers, nor PSFs that
schools / the number of junior high school students. infants and children often visit, such as community healthcare clinics
Compared with previous methods, the MGH3SFCA method takes and junior high schools, were distributed in high-density in Nanjing
residents' differentiated needs for PSFs into account, which better aligns Hexi Central Business District, so it could not satisfy the basic needs of
with the community life circle planning. To further verify its feasibility, residents. In the context of Nanjing's vigorous development of new

6
L. Song et al. Cities 144 (2024) 104637

Fig. 4. The spatial distribution of various PSFs.

districts, it is particularly important not only to pay attention to the daily For children (Fig. S3), educational facilities are the most frequently
needs of seniors and increase the vitality of the Old Town, but also to visited PSFs. The average value of accessibility to kindergartens, pri­
build adequate PSFs in new districts to improve residents' quality of life. mary schools and junior high schools was 0.02, 0.01, and 0.01, respec­
tively. Moreover, preschool children in 195 communities (46.54 %)
3.2. The matching level between supply and demand at the community could access kindergartens within 5 min, pupils in 292 communities
scale (69.69 %) could access primary schools within 10 min, and junior high
school students in 292 communities (69.69 %) could access junior high
To compare the differentiation among different types of PSFs, we schools within 15 min. In terms of spatial patterns, the spatial distri­
normalized the accessibility values by the min-max normalization bution of accessibility to kindergartens and primary schools was more
method and divided them into five levels (i.e., very high, high, medium, uneven than that to junior high schools. Distance has a significant
low, and very low) according to the natural breaks method. impact on accessibility due to the short service radius of kindergartens
For infants (Fig. S2), the supply-demand matching of PSFs fully ex­ and primary schools. Especially at the community scale, the community
poses the problems in the allocation of medical care facilities. The centroid was used to represent the whole community, and if there were a
average value of accessibility to community healthcare clinics and few communities with centroids close to kindergartens or primary
community healthcare centers was 0.01 and 0.02, respectively. schools, overall accessibility after normalization would be reduced.
Compared with healthcare centers, the number and service areas of For adults (Fig. S4), cultural and sports facilities are visited
healthcare clinics were smaller. Although the accessibility was high for frequently. The average value of accessibility to community cultural
communities close to healthcare clinics, 405 communities (97.12 %) had centers and community sports centers was 0.25 and 0.06, respectively,
an accessibility value of 0, which meant that infants living in these higher than other types of PSFs. Moreover, the number and service areas
communities could not access healthcare clinics within 5 min, resulting of these two types of PSFs were large, so cultural centers were able to
in a low average value of accessibility. As an essential part of healthy serve 408 communities (96.91 %), and sports centers were able to serve
community life circle, healthcare clinics can provide the most basic 354 communities (84.09 %). In terms of spatial patterns, the spatial
health services for residents and fundamental guarantees for responding distribution of accessibility to cultural centers and sports centers varied
to public health emergencies (Wang et al., 2020). In the future, it is greatly. The former was more balanced, while the spatial heterogeneity
necessary to add more community healthcare clinics in areas with lower of the latter was significant, with high-value areas appeared in the
levels of accessibility, such as Gulou and Xuanwu District. southwest of Jianye and northeast of Qixia District and low-value areas

7
L. Song et al. Cities 144 (2024) 104637

Fig. 5. The spatial distribution of different age groups.

in the Old Town. To advance the national fitness campaign, Nanjing has to the large number of seniors in these two districts. Nanjing is one of the
built a lot of community sports centers in recent years, but almost none earliest cities in China to provide home-based elderly care service
of the newly-built community sports centers are located in the Old (Zhang et al., 2019). In 2023, Nanjing Municipal People's Government
Town. Actions such as opening school playgrounds and building pocket reported that 96 % of seniors chose to aging in place. To address the
parks can be taken to improve the accessibility to sports facilities in the increasingly serious aging issue, Nanjing should promptly provide uni­
Old Town. versal access to home-based service centers for seniors and establish an
For seniors (Fig. S5), elderly care facilities play a crucial role in their elderly care service system based on home, supported by the commu­
daily life. The average value of accessibility to home-based service nity, supplemented by institutions and combined with medical care
centers and nursing homes was 0.03 and 0.02, respectively. Moreover, service.
seniors in 266 communities (63.48 %) could access home-based service To compare the difference among the four age groups, we utilized the
centers within 10 min, while only 206 communities (49.16 %) could entropy method to calculate the comprehensive accessibility for each
access nursing homes within 15 min. Although elderly care facilities age group. Fig. 6 showed that age groups with the same walking speed
were mainly distributed in Gulou and Qinhuai District, where the aging had a similar spatial pattern of comprehensive accessibility, indicating
rate had exceeded 20 %, the accessibility values were still not high due that walking speed had a great impact on the spatial pattern of

8
L. Song et al. Cities 144 (2024) 104637

Fig. 6. The supply-demand matching level of PSFs at the community scale.

comprehensive accessibility. Limited by walking speed, the average (community scale) to 96.06 % (building scale), and that to healthcare
value of comprehensive accessibility for children (0.04) and adults centers had also decreased from 68.82 % (community scale) to 65.97 %
(0.04) was higher than that for infants (0.02) and seniors (0.03). Spe­ (building scale). Compared with the accessibility at the community
cifically, for medical care facilities, the service coverage rates for infants scale, the accessibility at the building scale was improved. Nevertheless,
and seniors were 33.09 % and 32.94 %, respectively, while those for the problem that current medical care facilities in the main urban area of
children and adults were 64.44 % and 64.61 %, respectively (Fig. S6). Nanjing cannot meet the needs of residents is still knotty. With the
Moreover, comprehensive accessibility varied depending on the fre­ implement of the “three-child policy”, Nanjing may encounter more
quency of visits to PSFs for each age group. For instance, for shopping severe challenges. It is urgent to rationally allocate resources such as
facilities, as children usually did not go to fresh markets, the spatial medical care and early childhood care to encourage births.
patterns of accessibility for children and adults were different (Fig. S7). For children (Fig. S10), instead, the proportion of communities (or
buildings) inaccessible to kindergartens, primary schools and junior
high schools had increased from 53.46 %, 30.31 %, 30.31 % (community
3.3. The matching level between supply and demand at the building scale
scale) to 61.21 %, 35.20 %, 47.63 % (building scale), respectively. In
terms of spatial patterns, the spatial distribution of accessibility to pri­
As we mentioned before, the supply-demand matching of PSFs at the
mary schools at the building scale was more equitable than that at the
community scale is not always accurate due to the limited service scope
community scale. At the community scale, compared with junior high
of PSFs, especially for large communities. Therefore, we further
schools, the spatial heterogeneity of accessibility to primary schools was
analyzed the supply-demand matching level of PSFs at the building
more apparent, and low-value areas were more widely distributed.
scale.
However, at the building scale, the spatial pattern of primary school
For infants (Fig. S9), the proportion of communities (or buildings)
accessibility was similar to that of junior high school, and high-value
inaccessible to healthcare clinics had decreased from 97.12 %

9
L. Song et al. Cities 144 (2024) 104637

areas were significantly increased. Nanjing Municipal People's Govern­ For seniors (Fig. S12), the proportion of communities (or buildings)
ment has taken a series of effective actions in the last few years, such as inaccessible to home-based service centers had decreased from 36.52 %
developing public kindergartens, fostering exam-free enrollment and (community scale) to 32.05 % (building scale), and that to nursing
strengthening the allocation of educational resources in weak areas, but homes had also decreased from 50.84 % (community scale) to 46.26 %
still has far to go to achieve a new level of educational equity. (building scale). However, the spatial distribution of elderly care facil­
For adults (Fig. S11), the proportion of communities (or buildings) ities at the building scale was more uneven than that at the community
inaccessible to cultural centers and sports centers had also increased scale, and low-value areas of accessibility were more widely distributed.
from 3.09 %, 15.91 % (community scale) to 3.28 %, 16.79 % (building This phenomenon was mainly caused by the fact that a few home-based
scale), respectively. In terms of spatial patterns, the supply-demand service centers and nursing homes only provided services for a small
matching of PSFs at the building scale was worse than that at the number of seniors due to the inadequate construction of pedestrian road
community scale. For cultural centers, compared with the balanced networks, translating into extremely high accessibility values of some
distribution at the community scale, there were high accessibility clus­ buildings, but overall low accessibility levels.
ters in the Old Town at the building scale, similar to the kernel density As shown in Fig. 7, the accessibility at the building scale can fully
spatial pattern. Notably, the accessibility value was low in high popu­ reflect the differences within communities. Taking Xiangfangxincun
lation density areas, such as areas in the northwest corner of Qinhuai Community in Qinhuai District as an example, for infants, the west of
District. For sports centers, low-value areas were more extensive at the community was highly accessible, while the east was inaccessible.
building scale than at the community scale. Particularly, disturbed by Although age groups had a similar spatial pattern of comprehensive
extremely high accessibility values of a few communities, the accessi­ accessibility at community and building scales, the service coverage rate
bility to fresh markets was generally low at the community scale, while of PSFs was larger at the building scale. From community scale to
high-value areas were widespread at the building scale. building scale, the service coverage rate of PSFs had increased from

Fig. 7. The supply-demand matching level of PSFs at the building scale.

10
L. Song et al. Cities 144 (2024) 104637

33.09 % to 37.42 % for infants, from 98.33 % to 99.50 % for children,

from 98.34 % to 99.32 % for adults, and from 91.65 % to 95.64 % for
seniors. Specifically, for shopping facilities (Fig. S14), the spatial pat­
terns of accessibility for children and adults at the building scale were
different from those at the community scale. The average value of
accessibility for children (i.e., 0.11) was higher than that for adults (i.e.,
0.05) at the community scale, while the average value of accessibility for
adults (i.e., 0.07) was higher than that for children (i.e., 0.05) at the
building scale. One possible explanation was that the overall accessi­
bility to fresh markets was underestimated at the community scale due
to the interference of extremely high values, while at the building scale,
the accessibility to fresh markets was more balanced than that to su­
permarkets, so the accessibility to shopping facilities for adults was more
even than that for children.

4. Discussion

4.1. Comparison between the single-group and multi-group method

Fig. 8 showed that the two methods generated similar accessibility

patterns. The service coverage rate of community healthcare centers Fig. 9. Scatter-plot between estimates from single-group and multi-
from MH3SFCA (single-group) was 66.15 % (25, 840 to 39, 062), and group method.
that from MGH3SFCA (multi-group) was also high at 65.29 % (25, 505
to 39, 062). However, the average value of accessibility from MH3SFCA values. Significantly, MH3SFCA underestimated the accessibility in a
(i.e., 0.03) was higher than that from MGH3SFCA (i.e., 0.02). Further­ few communities. One possible explanation was that MH3SFCA assumed
more, we plotted a scatter plot (Fig. 9) to analyze the overestimation of that everyone had the same walking speed, but actually infants and
accessibility by MH3SFCA. It could be seen that the accessibility values seniors walked slower than children and adults and had fewer commu­
from MH3SFCA were mostly higher than those from MGH3SFCA. As the nity healthcare centers to access within 15 min, resulting in smaller
value increased, the tendency to overestimate accessibility became denominator of Eq. (5) and higher Rj and Ai values.
increasingly apparent. Overall, MH3SFCA ignored the differentiated needs of different age
In terms of spatial patterns, MH3SFCA overestimated the accessi­ groups, leading to the widespread overestimation of accessibility. In
bility to community healthcare centers especially in the Old Town. comparison, the proposed MGH3SFCA method could better reflect the
There were a large number of seniors who visited community healthcare inequity among regions.
centers more frequently than other age groups in the Old Town.
MH3SFCA assumed that everyone visited community healthcare centers
4.2. Quantification of differentiated needs of different resident groups
with the same frequency, while MGH3SFCA took into account the fact
that infants and seniors visited community healthcare centers more
Differentiated needs of residents play an important role in the
frequently than children and adults and generated larger demand,
accessibility to PSFs, so it is necessary to quantify the differentiated
resulting in larger denominator of Eq. (5) and lower Rj values. Moreover,
needs of different resident groups. Considering that the age structure is
seniors walked slower than children and adults and had fewer commu­
the main factor affecting the difference in residents' behavioral demands
nity healthcare centers to access within 15 min, resulting in lower Ai
(Li, 2017), we divided residents into four groups according to the age

Fig. 8. Comparison between the single-group and multi-group method.

11
L. Song et al. Cities 144 (2024) 104637

structure, and analyzed the travel distance and visit frequency charac­ infants and seniors), so their needs should be given in priority.
teristics of different age groups. In terms of travel distance, as commu­ Moreover, the accessibility results indirectly reflected residents'
nity life circle highlights the walking accessibility, we divided PSFs into needs. In places where high-frequency demand populations were
three categories (5 min, 10 min, and 15 min), and determined the travel more concentrated, the supply-to-demand ratio of PSFs and the
distance of different age groups according to their walking speed. accessibility value were lower. Therefore, more PSFs need to be
Nevertheless, the goals arbitrarily set by planners may not necessarily be added in areas with high-frequency demand populations.
acceptable to local residents (Logan et al., 2022). Further research is ii. Public participation: The core of community life circle planning is
needed to identify the maximum travel distance acceptable to local “people-oriented”, and the bottom-up allocation model domi­
residents. In terms of visit frequency, we divided visit frequency into low nated by residents is superior to the top-down allocation model
(0.2), medium (0.5) and high (0.8). It is essential to obtain accurate data dominated by the government (Cui et al., 2022). Therefore,
on how often residents visit PSFs. On the one hand, visit frequency can public participation in the planning is necessary. Community
be used in Eq. (5) as a coefficient to adjust the population and directly residents can be stimulated to fill out questionnaires to collect
reflect residents' differentiated needs (Cui et al., 2022). On the other information on the maximum travel distance they can tolerate
hand, it can be considered as a measure of the importance of PSFs to and the frequency and duration of visits to PSFs, which are
residents, and can be used to weight different types of PSFs in calcu­ important parameters for accessibility assessment. Based on the
lating comprehensive accessibility, which is more accurate than the urgent needs of local residents, the priority of allocating PSFs can
entropy method we used. be determined. Notably, vulnerable groups such as children, se­
The key to quantify residents' differentiated needs lies in the acqui­ niors and people with mobility impairment also need to be
sition of residents' travel behavior data. Collecting targeted question­ inspired to have a voice in community life circle planning, and
naires in local areas is one of the traditional ways to obtain travel appropriate ways of engaging them should be designed (Cordero-
behavior data of different resident groups. For example, McDonald Vinueza et al., 2023).
(2015) compared the travel behaviors of millennials with those of pre­ iii. Mixed land use: Mixed land use here means that every square
vious generations based on the data from the U.S. Department of meter can be used for different things (Moreno, 2020). Commu­
Transportation's National Household Travel Survey (NHTS), which is nity life circle planning encourages mixed land use to maximize
authoritative and scientific. Zhang et al. (2022) collected questionnaires available spaces and achieve close proximity between PSFs
to clarify how residents changed their travel behaviors to urban parks (Allam, Bibri, Chabaud, & Moreno, 2022). Not all buildings could
after the COVID-19 outbreak. Significantly, according to the results of access multi-type PSFs in the main urban area of Nanjing, and
Section 3.3, residents in the same community may face different prac­ medical care facilities were seriously inadequate to fulfill the
tical problems, so reasonable design and distribution of questionnaires needs of residents. Moreover, the available land was limited,
are very important. especially in the Old Town. Mixed land use may be one of the
In an era of personalized development, the needs of residents are important ways to address this issue. For instance, faced with the
increasingly diversified and differentiated (Liu, Meng, et al., 2022). problem that both the quantity and quality of medical care fa­
Residents with different genders, incomes, regions, and cultures have cilities and elderly care facilities need to be improved, high-
different needs for PSFs, making it more difficult to accurately quantify quality integrated elderly care and medical services for seniors
the differences in needs among different resident groups. With the can be boosted in the future. Additionally, Nanjing aims to build
development of big data such as taxi data, smart card data and mobile an international sports city and a model city for the national
phone signaling data, scholars have begun to investigate residents' fitness campaign, but sports facilities were insufficient in the Old
diversified needs by analyzing their spatiotemporal behaviors (Jing Town, where the utilization of school sports facilities was also
et al., 2023; Lee et al., 2023; Li et al., 2022; Zhang et al., 2023). For low, so school playgrounds can be fostered to be open to society,
example, Jing et al. (2023) identified patients' travel behaviors to hos­ especially on holidays and at night.
pitals based on taxi data. Nevertheless, there are still some obstacles to iv. Convenient transportation: A important finding of our study is
implementation, such as high costs and complex operation (Chai et al., that road networks have a great impact on accessibility. For
2020), and the actual needs of vulnerable groups such as seniors and example, since pedestrian road networks were not well-built, a
people with mobility impairment cannot be well investigated and few home-based service centers and nursing homes only provided
quantified. services for a small number of seniors, translating into extremely
high accessibility values of some buildings, but overall low
4.3. Implementation paths for 15-minute community life circle planning accessibility levels. In addition, in the main urban area of Nanj­
ing, it is difficult for residents to access multi-type PSFs within 15
In the process of rapid urbanization, urban diseases (Zhang et al., min only by walking, let alone in those cities with lower popu­
2021), such as traffic congestion (Kan et al., 2022), urban sprawl (Van lation and facility density. In the context of global decarburiza­
der Borght & Pallares Barbera, 2023) and air pollution (Yu et al., 2022), tion (Van der Borght & Pallares Barbera, 2023), we indeed need
have appeared in many cities, which go against people's ever-growing to reduce the use of private cars, but bicycles and even public
needs for a better life. It is urgent to improve urban living environ­ transport should be incorporated into community life circle, just
ment and residents' quality of life via urban planning. The important like walking (Tahmasbi et al., 2019). More safe, convenient and
value of community life circle planning is that it aligns with the concept green streets for pedestrians and cyclists can be created in cities
of “people-oriented” and adjusts the supply of PSFs according to resi­ (Nieuwenhuijsen, 2021), to improve the matching between the
dents' daily needs. We put forward some feasible suggestions for the supply and demand of PSFs.
construction of 15-minute community life circle as follows:
5. Conclusion
i. Targeted allocation: Although the service coverage rate of
comprehensive accessibility was high for the other three age In this study, we quantified the accessibility to PSFs for different age
groups except infants, not all buildings could access multi-type groups at community and building scales based on their differentiated
PSFs, so it is significant to rationally allocate PSFs in Nanjing. needs. The results showed that the supply-demand between PSFs and
Travel distance was the main factor affecting accessibility for different age groups was relatively mismatched in the main urban area
different age groups, and the supply and demand were more of Nanjing, and medical care facilities were seriously inadequate to
mismatched for the group with shorter travel distances (e.g., fulfill the needs of any age group. Comprehensive accessibility varied

12
L. Song et al. Cities 144 (2024) 104637

depending on the frequency of visits to PSFs for each age group. References
Although age groups had a similar spatial pattern of comprehensive
accessibility at community and building scales, the service coverage rate Allam, Z., Bibri, S. E., Chabaud, D., & Moreno, C. (2022). The theoretical, practical, and
technological foundations of the 15-minute city model: Proximity and its
of PSFs was larger at the building scale, and the assessment of supply- environmental, social and economic benefits for sustainability. Energies, 15(16),
demand matching of PSFs at the building scale was more accurate and Article 6042. [Link]
reliable. Allam, Z., Bibri, S. E., Jones, D. S., Chabaud, D., & Moreno, C. (2022). Unpacking the ‘15-
minute city’ via 6G, IoT, and Digital Twins: Towards a new narrative for increasing
The main contributions of this study were: (i) building an analytical urban efficiency, resilience, and sustainability. Sensors, 22(4), Article 1369. https://
framework of supply-demand matching of PSFs in community life circle [Link]/10.3390/s22041369
based on the actual needs of different age groups; (ii) extending the Burke, J., Aloma, R. G., Yu, F., & Kruguer, J. (2022). Geospatial analysis framework for
evaluating urban design typologies in relation with the 15-minute city standards.
accessibility assessment of PSFs from community scale to building scale, Journal of Business Research, 151, 651–667. [Link]
which was more accurate and reliable; and (iii) proposing the jbusres.2022.06.024
MGH3SFCA method to quantify the accessibility to PSFs for different age Calafiore, A., Dunning, R., Nurse, A., & Singleton, A. (2022). The 20-minute city: An
equity analysis of Liverpool City Region. Transportation Research Part D: Transport
groups by considering their behavioral demand differences.
and Environment, 102, Article 103111. [Link]
Some limitations also need to be acknowledged. Firstly, the quanti­ Chai, Y., Li, C., Xia, W., Wang, J., Zhang, X., & Sun, D. (2019). Study on the delineation
fication of the differentiated needs of different age groups was subjec­ model of urban community life circle: Based on Qinghe District in Haidian District,
tive. As mentioned in Section 4.2, questionnaires or big data, which Beijing. Urban Development Studies, 26(9), 1–8. [Link]
odical/csfzyj201909001.
accurately reflect residents' travel behaviors, has great potential to Chai, Y., Li, C., & Zhang, Y. (2020). A new time-geography research framework of
quantify residents' differentiated needs. Secondly, due to the lack of community life circle. Progress in Geography, 39(12), 1961–1971. [Link]
capacity information in POIs data, we had to assume that each facility 10.18306/dlkxjz.2020.12.001
Chen, Y., Chen, Z., Guo, D., Zhao, Z., Lin, T., & Zhang, C. (2022). Underground space use
has the same capacity. The accuracy and credibility of the results could of urban built-up areas in the central city of Nanjing: Insight based on a dynamic
be further improved by supplementing the capacity of various PSFs. population distribution. Underground Space, 7(5), 748–766. [Link]
Moreover, we only evaluated and analyzed the current status of PSFs in 10.1016/[Link].2021.12.006
Chen, Z., Li, P., Jin, Y., Jin, Y., Chen, J., Li, W., Song, X., Shibasaki, R., Chen, M., Yan, D.,
the main urban area of Nanjing, ignoring the dynamic changes in pop­ & Zhang, H. (2022). Using mobile phone big data to identify inequity of artificial
ulation and land use over time. Additionally, although we used building- light at night exposure: A case study in Tokyo. Cities, 128, Article 103803. https://
scale population data to support fine planning, it is difficult to identify [Link]/10.1016/[Link].2022.103803
Cheng, Z., Li, X., & Zhang, Q. (2023). Can new-type urbanization promote the green
all residential buildings due to the volume of building data (Zheng et al., intensive use of land? Journal of Environmental Management, 342, Article 118150.
2020). Especially in China, a residential building is usually in a certain [Link]
walled residential compound, where people enter and exit through the Cordero-Vinueza, V. A., Niekerk, F., & van Dijk, T. (2023). Making child-friendly cities: A
socio-spatial literature review. Cities, 137, Article 104248. [Link]
designated gates (Liu & Chai, 2015). How to accurately identify resi­
[Link].2023.104248
dential buildings and residents' trajectories is a key issue to study. Cui, X., Ma, L., Tao, T., & Zhang, W. (2022). Do the supply of and demand for rural public
Nonetheless, this study provides a new perspective and method for service facilities match? Assessment based on the perspective of rural residents.
supply-demand matching assessment of PSFs, which would contribute to Sustainable Cities and Society, 82, Article 103905. [Link]
scs.2022.103905
our understanding of the multidimensional relationships involved in Dadashpoor, H., Rostami, F., & Alizadeh, B. (2016). Is inequality in the distribution of
community life circle. urban facilities inequitable? Exploring a method for identifying spatial inequity in an
Iranian city. Cities, 52, 159–172. [Link]
Dai, D. (2010). Black residential segregation, disparities in spatial access to health care
CRediT authorship contribution statement facilities, and late-stage breast cancer diagnosis in metropolitan Detroit. Health &
Place, 16(5), 1038–1052. [Link]
Lu Song: Conceptualization, Methodology, Data processing, Writing - Dai, D. (2011). Racial/ethnic and socioeconomic disparities in urban green space
accessibility: Where to intervene? Landscape and Urban Planning, 102(4), 234–244.
Original draft. Xuesong Kong: Conceptualization, Supervision, Writing - [Link]
Review and editing, Funding acquisition. Peng Cheng: Methodology, Delamater, P. L. (2013). Spatial accessibility in suboptimally configured health care
Data processing, Writing - Review and editing. All authors contributed systems: A modified two-step floating catchment area (M2SFCA) metric. Health &
Place, 24, 30–43. [Link]
to the article and approved the submitted version. Dong, Y., Cheng, P., & Kong, X. (2022). Spatially explicit restructuring of rural
settlements: A dual-scale coupling approach. Journal of Rural Studies, 94, 239–249.
Declaration of competing interest [Link]
Dony, C. C., Delmelle, E. M., & Delmelle, E. C. (2015). Re-conceptualizing accessibility to
parks in multi-modal cities: A variable-width floating catchment area (VFCA)
The authors declare that they have no known competing financial method. Landscape and Urban Planning, 143, 90–99. [Link]
interests or personal relationships that could have appeared to influence landurbplan.2015.06.011
the work reported in this paper. Feng, J., & Yang, Z. (2015). Factors influencing travel behavior of urban elderly people in
Nanjing. Progress in Geography, 34(12), 1598–1608. [Link]
dlkxjz.2015.12.008
Data availability Guo, R., Li, Y., & Huang, M. (2019). Research on optimization strategy of walking
network in 15-minute community life circle of Harbin. Planners, 35(4), 18–24. https
://[Link]/periodical/ghs201904003.
Data will be made available on request. Han, Z., Cui, C., Kong, Y., Li, Q., Chen, Y., & Chen, X. (2023). Improving educational
equity by maximizing service coverage in rural Changyuan, China: An evaluation-
Acknowledgment optimization-validation framework based on spatial accessibility to schools. Applied
Geography, 152, Article 102891. [Link]
Hu, L., Zhao, C., Wang, M., Su, S., Weng, M., & Wang, W. (2020). Dynamic healthy food
This research was funded by the National Social Science Fund of accessibility in a rapidly urbanizing metropolitan area: Socioeconomic inequality
China (23BGL267). and relative contribution of local factors. Cities, 105, Article 102819. [Link]
org/10.1016/[Link].2020.102819
Hu, S., Song, W., Li, C., & Lu, J. (2020). A multi-mode Gaussian-based two-step floating
Appendix A. Supplementary data catchment area method for measuring accessibility of urban parks. Cities, 105,
Article 102815. [Link]
Institute for Transportation and Development Policy. (2022). Access and babies,
Supplementary data to this article can be found online at [Link]
toddlers, and their caregivers. [Link]
org/10.1016/[Link].2023.104637. es-toddlers-and-their-caregivers/.
Jing, C., Zhou, W., Qian, Y., Zheng, Z., Wang, J., & Yu, W. (2023). Trajectory big data
reveals spatial disparity of healthcare accessibility at the residential neighborhood
scale. Cities, 133, Article 104127. [Link]

13
L. Song et al. Cities 144 (2024) 104637

Jörg, R., & Haldimann, L. (2023). MHV3SFCA: A new measure to capture the spatial Ministry of Natural Resources of the People'’s Republic of China. (2021). Spatial
accessibility of health care systems. Health & Place, 79, Article 102974. [Link] planning guidance to community life unit. [Link]
org/10.1016/[Link].2023.102974 e/read/21d2d1d71032b84e847e2baeb6aaf39c.
Kan, Z., Kwan, M.-P., Liu, D., Tang, L., Chen, Y., & Fang, M. (2022). Assessing individual Moreno, C. (2020). The 15-minute city. [Link]
activity-related exposures to traffic congestion using GPS trajectory data. Journal of e_15_minute_city.
Transport Geography, 98, Article 103240. [Link] Moreno, C., Allam, Z., Chabaud, D., Gall, C., & Pratlong, F. (2021). Introducing the “15-
jtrangeo.2021.103240 Minute City”: Sustainability, resilience and place identity in future post-pandemic
Kesarovski, T., & Hernández-Palacio, F. (2022). Time, the other dimension of urban cities. Smart Cities, 4(1), 93–111. [Link]
form: Measuring the relationship between urban density and accessibility to grocery Nanjing Municipal Bureau of Planning and Natural Resources. (2023). Nanjing planning
shops in the 10-minute city. Environment and Planning B: Urban Analytics and City guidance of 15-minute community life circle. [Link]
Science, 50(1), 44–59. [Link] 2306/t20230608_3931597.html.
Khavarian-Garmsir, A. R., Sharifi, A., & Sadeghi, A. (2023). The 15-minute city: Urban Nanjing Municipal People's Government. (2023). Nanjing planning standard for public
planning and design efforts toward creating sustainable neighborhoods. Cities, 132, facilities. [Link]
Article 104101. [Link] Nieuwenhuijsen, M. J. (2021). New urban models for more sustainable, liveable and
Langford, M., Higgs, G., & Fry, R. (2016). Multi-modal two-step floating catchment area healthier cities post covid19; Reducing air pollution, noise and heat island effects
analysis of primary health care accessibility. Health & Place, 38, 70–81. [Link] and increasing green space and physical activity. Environment International, 157,
org/10.1016/[Link].2015.11.007 Article 106850. [Link]
Lee, S., Ko, E., Jang, K., & Kim, S. (2023). Understanding individual-level travel behavior Oostenbach, L. H., Lamb, K. E., Crawford, D., Timperio, A., & Thornton, L. E. (2023).
changes due to COVID-19: Trip frequency, trip regularity, and trip distance. Cities, Cross-sectional associations between work and commute hours with frequency and
135, Article 104223. [Link] location of food outlet visits: The moderating role of 20-minute neighbourhoods.
Li, C., Xia, W., & Chai, Y. (2021). Delineation of an urban community life circle based on Cities, 137, Article 104345. [Link]
a machine-learning estimation of spatiotemporal behavioral demand. Chinese Shanghai Planning and Land Resources Administration. (2016). Shanghai planning
Geographical Science, 31(1), 27–40. [Link] guidance of 15-minute community-life circle. [Link]
Li, M. (2017). The planning strategies of a 15-minute community life circle based on ss/201609/[Link].
behaviors of residents. Urban Planning Forum, 2017(1), 111–118. [Link] Song, G., He, X., Kong, Y., Li, K., Song, H., Zhai, S., & Luo, J. (2022). Improving the
10.16361/[Link].201701014 spatial accessibility of community-level healthcare service toward the ‘15-minute
Li, Z., Xiong, G., Wei, Z., Zhang, Y., Zheng, M., Liu, X., Tarkoma, S., Huang, M., Lv, Y., & city’ goal in China. ISPRS International Journal of Geo-Information, 11(8), Article 436.
Wu, C. (2022). Trip purposes mining from mobile signaling data. IEEE Transactions [Link]
on Intelligent Transportation Systems, 23(8), 13190–13202. [Link] Subal, J., Paal, P., & Krisp, J. M. (2021). Quantifying spatial accessibility of general
tits.2021.3121551 practitioners by applying a modified huff three-step floating catchment area
Lin, Y., Zhou, Y., Lin, M., Wu, S., & Li, B. (2021). Exploring the disparities in park (MH3SFCA) method. International Journal of Health Geographics, 20, Article 9.
accessibility through mobile phone data: Evidence from Fuzhou of China. Journal of [Link]
Environmental Management, 281, Article 111849. [Link] Sun, D., & Chai, Y. (2018). Review and inspiration of researches on sphere of life in
jenvman.2020.111849 Japan. Urbanism and Architecture, 2018(36), 13–16. [Link]
Liu, J., Meng, B., Yang, M., Peng, X., Zhan, D., & Zhi, G. (2022). Quantifying spatial [Link].2018.36.003
disparities and influencing factors of home, work, and activity space separation in Sun, D., Chai, Y., & Zhang, Y. (2016). The definition and measurement of community life
Beijing. Habitat International, 126, Article 102621. [Link] circle: A case study of Qinghe Area in Beijing. Urban Development Studies, 23(9), 1–9.
habitatint.2022.102621 [Link]
Liu, K., Xue, Y., Chen, Z., Miao, Y., & Shi, J. (2022). Economic spatial structure of China's Tahmasbi, B., Mansourianfar, M. H., Haghshenas, H., & Kim, I. (2019). Multimodal
urban agglomerations: Regional differences, distribution dynamics, and accessibility-based equity assessment of urban public facilities distribution.
convergence. Sustainable Cities and Society, 87, Article 104253. [Link] Sustainable Cities and Society, 49, Article 101633. [Link]
10.1016/[Link].2022.104253 scs.2019.101633
Liu, P., & Zhu, B. (2022). Temporal-spatial evolution of green total factor productivity in Tao, Z., & Cheng, Y. (2016). Research progress of the two-step floating catchment area
China's coastal cities under carbon emission constraints. Sustainable Cities and method and extensions. Progress in Geography, 35(5), 589–599. [Link]
Society, 87, Article 104231. [Link] 10.18306/dlkxjz.2016.05.006
Liu, Q., Qian, Z., Huang, D., & Zhou, X. (2020). The characteristics of spatial pattern United Nations. (2015). Transforming our world: The 2030 agenda for sustainable
evolution and the trend of the 15-minute life circle. Urban Planning Forum, 2020(6), development. [Link]
94–101. [Link] Van der Borght, R., & Pallares Barbera, M. (2023). How urban spatial expansion
Liu, T., & Chai, Y. (2015). Daily life circle reconstruction: A scheme for sustainable influences CO2 emissions in Latin American countries. Cities, 139, Article 104389.
development in urban China. Habitat International, 50, 250–260. [Link] [Link]
10.1016/[Link].2015.08.038 Wan, N., Zou, B., & Sternberg, T. (2012). A three-step floating catchment area method for
Liu, W., Zheng, S., Hu, X., Wu, Z., Chen, S., Huang, Z., & Zhang, W. (2022). Effects of analyzing spatial access to health services. International Journal of Geographical
spatial scale on the built environments of community life circles providing health Information Science, 26(6), 1073–1089. [Link]
functions and services. Building and Environment, 223, Article 109492. [Link] 13658816.2011.624987
org/10.1016/[Link].2022.109492 Wang, L., Li, X., & Yang, X. (2020). Health in 15-minute life sphere: Community response
Logan, T. M., Hobbs, M. H., Conrow, L. C., Reid, N. L., Young, R. A., & Anderson, M. J. to public health emergency. Planners, 36(6), 102–106. [Link]
(2022). The x-minute city: Measuring the 10, 15, 20-minute city and an evaluation of cn/periodical/ghs202006021.
its use for sustainable urban design. Cities, 131, Article 103924. [Link] Wang, Y., Liu, Y., Xing, L., & Zhang, Z. (2021). An improved accessibility-based model to
10.1016/[Link].2022.103924 evaluate educational equity: A case study in the city of Wuhan. ISPRS International
Luo, J. (2014). Integrating the Huff Model and floating catchment area methods to Journal of Geo-Information, 10(7), Article 458. [Link]
analyze spatial access to healthcare services. Transactions in GIS, 18(3), 436–448. ijgi10070458
[Link] Weng, M., Ding, N., Li, J., Jin, X., Xiao, H., He, Z., & Su, S. (2019). The 15-minute
Luo, W., & Qi, Y. (2009). An enhanced two-step floating catchment area (E2SFCA) walkable neighborhoods: Measurement, social inequalities and implications for
method for measuring spatial accessibility to primary care physicians. Health & building healthy communities in urban China. Journal of Transport and Health, 13,
Place, 15(4), 1100–1107. [Link] 259–273. [Link]
Luo, W., & Wang, F. (2003). Measures of spatial accessibility to health care in a GIS Wu, J., Chen, H., Wang, H., He, Q., & Zhou, K. (2020). Will the opening community
environment: Synthesis and a case study in the Chicago region. Environment and policy improve the equity of green accessibility and in what ways? — Response
Planning. B, Planning & Design, 30(6), 865–884. [Link] based on a 2-step floating catchment area method and genetic algorithm. Journal of
Luo, W., & Whippo, T. (2012). Variable catchment sizes for the two-step floating Cleaner Production, 263, Article 121454. [Link]
catchment area (2SFCA) method. Health & Place, 18(4), 789–795. [Link] jclepro.2020.121454
10.1016/[Link].2012.04.002 Xia, N., Cheng, L., Chen, S., Wei, X., Zong, W., & Li, M. (2018). Accessibility based on
Luo, X., Zhang, W., & Chai, Y. (2022). Research on threshold effects of built environment Gravity-Radiation model and Google Maps API: A case study in Australia. Journal of
settings in 15-minute life-circles. Geographical Research, 41(8), 2155–2170. https:// Transport Geography, 72, 178–190. [Link]
[Link]/10.11821/dlyj020210851 Xia, Y., Chen, H., Zuo, C., & Zhang, N. (2022). The impact of traffic on equality of urban
Mao, L., & Nekorchuk, D. (2013). Measuring spatial accessibility to healthcare for healthcare service accessibility: A case study in Wuhan, China. Sustainable Cities and
populations with multiple transportation modes. Health & Place, 24, 115–122. Society, 86, Article 104130. [Link]
[Link] Xiao, Y., Wang, D., & Fang, J. (2019). Exploring the disparities in park access through
McDonald, N. C. (2015). Are millennials really the “go-nowhere” generation? Journal of mobile phone data: Evidence from Shanghai, China. Landscape and Urban Planning,
the American Planning Association, 81(2), 90–103. [Link] 181, 80–91. [Link]
01944363.2015.1057196 Xiao, Z., Chai, Y., & Zhang, Y. (2014). Overseas life circle planning and practice. Planners,
McGrail, M. R., & Humphreys, J. S. (2014). Measuring spatial accessibility to primary 30(10), 89–95. [Link]
health care services: Utilising dynamic catchment sizes. Applied Geography, 54, Yu, P., Chen, Y., Xu, Q., Zhang, S., Yung, E. H. K., & Chan, E. H. W. (2022). Embedding of
182–188. [Link] spatial equity in a rapidly urbanising area: Walkability and air pollution exposure.
Ministry of Housing and Urban-Rural Development of the People'’s Republic of China. Cities, 131, Article 103942. [Link]
(2018). Standard for urban residential area planning and design. [Link]
[Link]/gongkai/zhengce/zhengcefilelib/201811/20181130_238590.html.

14
L. Song et al. Cities 144 (2024) 104637

Yu, Y. (2019). From traditional residential area planning to neighborhood life circle Zhao, P., Luo, J., & Hu, H. (2021). Spatial match between residents' daily life circle and
planning. City Planning Review, 43(5), 17–22. [Link] public service facilities using big data analytics: A case of Beijing. Progress in
dical/csgh201905004. Geography, 40(4), 541–553. [Link]
Zhang, F., Li, D., Ahrentzen, S., & Zhang, J. (2019). Assessing spatial disparities of Zhao, P., Luo, J., & Hu, H. (2022). Character of the elderly's life circle and public service
accessibility to community-based service resources for Chinese older adults based on facilities configuration by using big data: A case of Beijing. Scientia Geographica
travel behavior: A city-wide study of Nanjing, China. Habitat International, 88, Article Sinica, 42(7), 1176–1186. [Link]
101984. [Link] Zheng, Z., Shen, W., Li, Y., Qin, Y., & Wang, L. (2020). Spatial equity of park green space
Zhang, S., Luan, H., Zhen, F., Kong, Y., & Xi, G. (2023). Does online food delivery using KD2SFCA and web map API: A case study of Zhengzhou, China. Applied
improve the equity of food accessibility? A case study of Nanjing, China. Journal of Geography, 123, Article 102310. [Link]
Transport Geography, 107, Article 103516. [Link] Zhong, T., Lu, G., Zhong, X., Tang, H., & Ye, Y. (2020). Measuring human-scale living
jtrangeo.2022.103516 convenience through multi-sourced urban data and a geodesign approach: Buildings
Zhang, W., Cao, J., He, J., & Chen, L. (2021). City health examination in China: A as analytical units. Sustainability, 12(11), Article 4712. [Link]
methodology and empirical study. Chinese Geographical Science, 31(6), 951–965. su12114712
[Link] Zou, S., Zhang, S., & Zhen, F. (2021). Measurement of community daily activity space
Zhang, W., Li, S., Gao, Y., Liu, W., Jiao, Y., Zeng, C., Gao, L., & Wang, T. (2022). Travel and influencing factors of vitality based on residents’spatiotemporal behavior:
changes and equitable access to urban parks in the post COVID-19 pandemic period: Taking Shazhou and Nanyuan streets in Nanjing as examples. Progress in Geography,
Evidence from Wuhan, China. Journal of Environmental Management, 304, Article 40(4), 580–596. [Link]
114217. [Link]

15