How Urban Block Form Affects the Vitality of the Catering Industry: Evidence from Jinan, China

Abstract

Existing research underscores the significance of block form in fostering urban vitality. However, there is a dearth of evidence on its influence on the catering industry’s vitality. Additionally, current studies investigating the correlation between block form and urban vitality have frequently neglected disparities among various types of blocks with dominant functions. We employ a multi-scale geographically weighted regression and a geographic detector to elucidate the effects of block form and the heterogeneity of blocks with dominant functions on the catering industry’s vitality. Our findings suggest that the suitable block form can mitigate the catering industry’s reliance on factors such as the regional population and public transportation infrastructure, to a certain degree. High-rise buildings and irregular block plans positively influence the catering industry’s vitality, while the effects of block area, building density, and functional density display considerable spatial heterogeneity. Commercial blocks are most influenced by block form, whereas public service blocks are least affected. The methodology of this study can be replicated globally to guide urban planners in judiciously allocating commercial facilities, based on the varying spatial form requirements of different blocks, thereby fostering a vibrant city.

1. Introduction

With the profound transformation of urban society, urban vitality has become an important topic in urban geography, planning, and economics [1,2]. Following Jacobs’s idea of urban diversity and Gehl’s theory of street space vitality, many urban renewal and redevelopment plans have started advocating for the construction of vibrant and diverse urban spaces [3,4]. It is therefore necessary to explore the diversification of urban vitality applications, based on a deep understanding of the connotations of urban vitality, to efficiently use resources and tap growth potential [5]. In-depth exploration of urban vitality and its influencing factors will enhance our comprehension of sustainable urban development and strategies to enhance residents’ quality of life [6].

Recently, the consumer economy has progressively transitioned from products to services. The substantial demand created by consumers has emerged as a pivotal driving force for both economic and social development [7]. The catering industry, a vital component of the urban service sector, has exhibited robust development potential and evolved into a wellspring of urban vitality [5].

Blocks are the most basic physical units of urban spaces and are places where residents most frequently conduct daily activities. Vibrant blocks not only enhance the spatial quality of cities, but also greatly affect residents’ quality of life and comfort [8]. However, because of the continuous impact of global public health events, such as the COVID-19 epidemic, many large cities in China have experienced shortcomings in block planning and services. A consensus has been reached regarding the creation of vibrant urban spaces based on human needs [9,10]. Simultaneously, with Chinese cities transitioning from rapid expansion to high-quality development, the scientific and efficient use of urban space and the full exploitation of economic growth potential have become key factors for achieving high-quality urban development [11,12]. In this context, exploring the mechanisms and effects of urban block forms on the catering industry’s vitality is of great significance, both for improving the efficiency of urban space use and for fully releasing economic development potential and vitality. Subsequently, this study considers the main urban area of Jinan, China, as the case study area. It also uses social network commentary and block form data to reveal the impact mechanism of the block form of different dominant functions on the catering industry’s vitality, thereby providing a theoretical basis for block planning construction and a reasonable layout for catering outlets.

2. Block Form, Urban Vitality, and Urban Catering Industry’s Vitality

2.1. Role of Block Form in the Study of Urban Vitality

The study of urban vitality originated quite early, with research focusing primarily on the concept of urban vitality and its influencing factors. Jacobs introduced the concept of urban vitality, believing that the intricate interplay between human activities and living spaces contributes to the diversity of urban life, making cities vibrant, and that the degree of crowd gathering represents urban vitality [3]. Lynch proposed that a vibrant city should be able to meet the various needs of residents, who should be able to conduct diverse activities in a safe living environment [13]. Furthermore, in “Life Between Buildings”, Gehl emphasized that individuals and their activities form the essence of urban public spaces’ vitality—that is, the selective and social activities that occur in the space, rather than necessary activities such as commuting [14].

Drawing upon earlier research, Montgomery presented the constituent elements of urban vitality: activity, interaction, and diversity [15]. Regarding the factors influencing the generation of vitality, owing to the urban shrinkage phenomenon in the western city transformation process, the micro-level architectural environment has become an significant factor in the exploration of urban vitality [16,17,18,19]. For example, Mahmoudi Farahani et al. found that differences in factors like land-use structure, density, block size, and accessibility affect urban vitality [20]. Bruneckiene et al. examined Kaunas City in Lithuania and evaluated the connection between the community and enterprises, using questionnaire surveys, spatial syntax analysis, and other methods [21]. They believed that community vitality was a major factor for the successful development of cities. Lopes and Camanho proposed that the integration of block environmental elements and the improvement of accessibility to specific facilities are important ways to create urban vitality [22]. As early as the commencement of the 20th century, researchers started focusing on the relationship between blocks and urban vitality. In 1933, the International Congress of Modern Architecture passed the “Athens Charter”, proposing the “four major functional areas of living, working, recreation, and transportation”. This efficiency-oriented programmatic document changed the traditional planning form of urban blocks and unveiled the hidden dangers for urban spatial vitality. The urban problems caused by functional zoning became increasingly prominent after the mid-20th century, when theories and practices of urban planning began to criticize and reflect on this mechanical functionalism, and people gradually realized that cities are complex mega-systems.

In the late 1980s, scholars began to re-examine the significance of street spaces for urban vitality and explore block planning methods that considered both material conditions and spiritual needs. New urbanism, sustainable development, smart growth, and other ideas gradually started emerging [15,23]. Today, many scholars advocate the construction of open, mixed-use, and vibrant urban blocks. Field studies, expert scoring, and multisource location service data have been employed to investigate the correlation between block form and urban vitality [24]. For example, Liu et al. focused on Zhoujiadu Street in Shanghai, built a quantitative evaluation indicator system and used a multivariate linear regression model [25]. They found that social function density, social function mix, distance from the nearest subway station, greening landscape, and other factors exert significant influences on urban vitality. Moreover, Wu et al. employed mobile positioning data to measure the block vitality of Nanjing West Road and surrounding areas in Shanghai [26]. They scrutinized the correlation between block vitality and the urban built environment, concluding that the design of external block spaces plays a pivotal role in sustaining vitality. Other scholars have explored the impact of three-dimensional built environment factors, such as building coverage ratio and building density, on urban sustainability [27,28].

2.2. From Urban Vitality to the Urban Catering Industry’s Vitality

With the advent of the consumer era, consumption spaces have become important components of urban spaces. As the most common places in urban consumption spaces, catering spaces have—to some extent—demonstrated urban vitality and shaped cities’ image. The catering industry, as a form of consumption space, reflects a region’s cultural foundation and lifestyle [29]. The distribution and differentiation of urban catering scales and types also signify the differentiation of social spaces [30]. As a carrier of “human–land” activities, a city is not only a cold geometric space; it should also be a warmer space that carries rich human emotional experiences. Against this background, geography and urban and rural planning studies have focused more on human behavior and emotions in space, with a shift toward “emotional turn” and “humanistic” thinking. Scholars, mainly in the postmodern context, have metaphorized catering space as a dimension of urban characteristics to enrich the understanding of human–land relations. For example, Neal used a two-step cluster analysis and divided 243 American cities into three types based on the number and type of restaurants each city has: “fine dining oasis”, “McDonaldized oasis”, and “dining desert” [31]. Schiff used data from more than 127,000 restaurants in 726 cities across the United States to measure urban diversity and explore the impact of population size and density on the diversity of urban catering establishments [32]. Meanwhile, Dock et al. used three various gravity model formulations to assess the geographical advantages and competitive capabilities of restaurants in Jefferson County, Kentucky, showing the correlation between restaurant market potential and spatial interactions with customer traffic [33]. With the widespread application of mobile Internet and information technology in the catering industry, its organizational modes and spatial distribution patterns have changed [34]. The integration of online platforms and traditional catering has spawned organizational modes, such as dine-in, take-out, and group buying. Offline catering spaces are mapped to the level of online spaces, while online catering spaces are rooted in offline spaces, enhancing their spatial stickiness. Additionally, the emergence of consumer review websites has provided people with a convenient platform for posting comments. The number of reviews for catering outlets usually far exceeds that for other commercial facilities [35], which may be because individuals tend to comment more on food facilities following their visits [36].

2.3. Research Review and Innovation

In summary, there is an extensive literature on block form, urban vitality, and urban catering spaces. However, owing to the many current uncertainties in urban development, the diversification of the application of urban vitality theory to solve emerging problems remains a significant challenge. Existing research has studied the human–environment relationship in urban catering spaces from two different analytical perspectives: real space and network information. Nonetheless, we must consider human activities, emotions, and urban spatial elements from a unified perspective of real space and network information to further enrich our understanding of catering spaces’ vitality. However, there has been little quantitative research on the urban catering industry’s vitality and its relationship with block forms.

Consequently, we attempt to enrich relevant research from three perspectives. First, based on the urban vitality research perspective, we define the connotation and evaluation system of the catering industry’s vitality from two dimensions: vitality intensity, and quality. Thereby, we enrich theoretical research on urban vitality and catering geography. Second, we consider the homogeneity and heterogeneity between blocks with different leading functions to better identify the driving forces of spatial differentiation in the catering industry. Third, we integrate the results of a multi-scale geographical weighted regression, geographical explorer, and other model analyses—especially comparing the effects of single driving factors and the interaction between two factors—to more comprehensively explain the influence mechanism of block form on the catering industry’s vitality. This study deepens our understanding of the correlation between block form and urban vitality, offering insights for block planning and construction.

3. Materials and Methodology

3.1. Study Area

Jinan is situated in the central and western regions of Shandong Province, China, positioned between the geographical coordinates of 36°01′ to 37°32′ north latitude and 116°11′ to 117°58′ east longitude. It is located south of Mount Tai and north of the Yellow River and has a sloping terrain elevated in the southern regions and lower in the northern areas. The area comprises low mountains and hills, sloping plains in front of the mountains, and alluvial plains near the Yellow River. Owing to geographical constraints, the old city of Jinan extends in an east–west orientation, creating a narrow and elongated urban spatial form. In its capacity as the capital of Shandong, Jinan serves as the political, economic, and cultural hub of the province. Its catering industry has developed significantly in China, in terms of market size, industrial structure, and variety. The study was conducted in the primary urban region of Jinan (Figure 1), encompassing an approximate area of 197.5 km². It constitutes a significant focal point regarding both population density and economic endeavors within Jinan City, rendering it conducive for research pertaining to urban block formations and the catering industry’s vitality.

3.2. Methodology

3.2.1. Measurement Method of the Catering Industry’s Vitality

Combining the connotations and measurement methods of urban vitality [37,38], we must consider not only the intensity of the interaction between people and the catering space, but also people’s feelings during the interaction to characterize the catering industry’s vitality. In the data on catering outlets, the number of reviews represents the popularity of a restaurant, and the average per capita consumption reflects customers’ willingness to consume. The main purpose of dining out is to satisfy one’s taste buds, so the taste of the restaurant’s food is an important factor for attracting consumers; a warm and friendly service can make diners feel welcome and cared for, thereby winning their favor and ensuring better word-of-mouth and income for the restaurant; and an elegant environment can create a comfortable and relaxed atmosphere, allowing diners to better enjoy food and social interaction. Contrarily, a bad atmosphere may make diners feel uncomfortable, affecting their dining experience. Therefore, the number of comments and the average consumption per capita were selected as indicators reflecting the intensity of the catering industry’s vitality, while the taste of food, atmosphere, and service scores were selected as indicators reflecting the quality of the catering industry’s vitality (

Table 1. Indicator system and descriptive statistics of the catering industry’s vitality.

Primary Index	Secondary Index	Indicator Attribute	Avg	SD	Max	Min
Vitality intensity	Number of comments	+	220.76	554.23	15,826.00	1.00
	Average consumption per capita	+	52.70	48.59	1278.00	2.00
Vitality quality	Taste score	+	4.03	0.40	4.96	3.08
	Atmosphere score	+	4.01	0.43	4.95	2.91
	Service score	+	4.01	0.41	4.97	2.92

Note: SD = standard deviation.

). The entropy method was used to measure catering outlets’ vitality [39]. It should be noted that the catering industry’s vitality in a block is the sum of the vitality values of all catering outlets within the block, while the “catering industry’s vitality”, referred to by the explained variable, is the catering industry’s vitality per unit area of the block.

3.2.2. Measurement Methods of the Urban Block Forms

According to the literature [40], block form indicators can generally be categorized into three types (

Table 2. Potential block form indicators affecting the catering industry’s vitality and their computing methods.

Primary Index	Secondary Index	Method of Calculation	Formula Meaning
Block plan form	Area	-	-
	Form ratio	A/L2	A is the area of the block, L is the length of the longest axis of the block.
	Circularity compactness	4A/P2	A is the area of the block, P is the perimeter of the block.
Block building form	Average floor number	$\sum N_i/n$	n represents the number of buildings in the block, $N_i$ represents the number of floors in building i.
	Floor area ratio	Q/S	Q is the total above-ground building area of the block, S is the base area of the block’s buildings.
	Building density	S/A	S is the base area of the block’s buildings, A is the area of the block.
Block functional form	Functional mix index	−$\sum \left(P_i\times \ln P_i\right)$	$P_i$ represents the proportion of the ith category of POI in the block.
	Functional density	n/A	n is the number of POIs in the block; A is the area of the block.

Note: POI = points of interest.

): (1) The plan form is measured based on the two-dimensional plan of the blocks, with common indicators including block area, form ratio, and circularity compactness. (2) Architectural form indicators are based on urban building information, analyzing the characteristics of block forms in the three-dimensional space, with area, floor ratio, building density, and average floor number as the most common variables. (3) Functional form indicators are based on points of interest (POI) data, considering blocks from the perspective of facilities and places, with common functional form indicators including dominant block function, functional density, and functional diversity.

First, potential block form indicators that may affect the catering industry’s vitality were selected. Among them, block area was measured using the “Calculate Geometry” tool in ArcGIS. The dominant block function was determined based on the main POI type within the blocks; if a certain type of POI accounted for more than 50% of all POIs, then the functional type of this POI was designated as the dominant block function, and blocks without a dominant function were defined as mixed functional blocks. The calculation methods for the other block form indicators are listed in Table 2. Finally, the SPSS 24 software stepwise regression method was used for independent variable selection, combined with significance and R² change, and the area, form ratio, average floor number, building density, and functional density were obtained as the five core explanatory variables.

3.2.3. Multi-Scale Geographically Weighted Regression

Following to the First Law of Geography, the proximity of two entities correlates with the strength of their connection. The geographically weighted regression (GWR) assigns higher weights to the observed values within the sample point’s neighborhood, reflecting the natural connection between things, that is, embedding the spatial structure into the linear regression model [41]. The structural equation for a classical GWR is as follows:

$$Y_i={\beta }_0\left(u_i,v_i\right)+{{\displaystyle \sum }}_k{\beta }_k\left(u_i,v_i\right)X_{ik}+{\epsilon }_i$$

where $Y_i$ is the dependent variable at point I; ${\beta }_0\left(u_i,v_i\right)$ is the intercept; ${\beta }_k\left(u_i,v_i\right)$ is the value of the continuous function $\beta \left(u,i\right)$ at point $\left(u_i,v_i\right)$; $X_{ik}$ is the value of the kth predictor variable at point I; and ${\epsilon }_i$ is the residual.

In contrast to conventional regression models like ordinary least squares, the classic GWR addresses the spatial heterogeneity of influencing factors to a certain extent by employing local regression. However, the use of a single kernel function and bandwidth to compute the weights leads to same-scale features for the spatial variations in all parameter estimates. Contrarily, the multi-scale geographically weighted regression (MGWR) derives each regression coefficient from local regression. Bandwidth, with its specificity, explains the ‘spatial scale effect’ of socioeconomic phenomena [42]. The linear regression equation for MGWR is as follows:

$$Y_i={{\displaystyle \sum }}_{k=1}^q{\beta }_0{X}_{ik}+{{\displaystyle \sum }}_{k=q+!}^p{\beta }_{bwk}\left(u_i,v_i\right)X_{ik}+{\epsilon }_i$$

where bwk denotes the bandwidth employed as the regression coefficient for the kth variable. This study utilized the Gaussian function method as a weight mechanism, which is formulated as follows:

$$W_{ij}=exp\left(-{\left(d_{ij}/b\right)}^2\right)$$

where $W_{ij}$ denotes the weight influence between dining locations i and j, b signifies the bandwidth. A larger bandwidth value results in a slower decay of weight influence with increasing distance $d_{ij}$. Conversely, the weight influence decays faster with an increase in distance $d_{ij}$, and the value of bandwidth b depends on the number of neighboring features or distance range. The Akaike information criterion (AIC) was used to optimize the bandwidth, and its formula is as follows:

$$AIC=2n\ln \left(\sigma \right)+n\ln \left(2\pi \right)+n\frac{n+tr\left(S\right)}{n-2-tr\left(S\right)}$$

in this equation, the optimal bandwidth for the model is determined by the one corresponding to the minimum AIC. Furthermore, n represents the number of sampling points; $tr\left(S\right)$ represents a trace of the bandwidth function S matrix in the MGWR model; and $\sigma$ represents the standard deviation of the error term estimation.

To mitigate the impact of other factors on the dependent variable, five factors were selected as control variables from the statistical and analytical attributes of the block, as follows: population density, bus stop density, betweenness centrality, distance to the nearest business district, and distance to the city center. There are several reasons for selecting these variables, as follows: People serve as both the creators and participants in activities, and they form the main body and foundation of urban vitality. Therefore, places with higher population densities may have a stronger catering industry vitality. Urban transportation is an important foundation for the successful operation of the catering industry. An efficient transportation system can not only increase the accessibility of restaurant facilities, but also help save consumer travel costs and improve the efficiency and impact range of restaurant facilities. Therefore, bus stop density and betweenness centrality were selected to reflect the impact of transportation on the restaurant industry’s vitality. The urban network analysis tool in ArcGIS was used to measure betweenness centrality [43]. Betweenness centrality suggests the shortest path connecting any two nodes in the transportation network, highlighting that paths passing through a specific area exhibit enhanced centrality [44]. The following formula was used:

$$C_i^B=\frac{1}{\left(N-1\right)\left(N-2\right)}{{\displaystyle \sum }}_{j=1,k=1,j\ne k\ne i}^N\frac{n_{jk}\left(i\right)}{n_{jk}}$$

where $C_i^B$ represents the betweenness centrality of node i, N stands for the total number of nodes, n_jk represents the number of shortest paths from node j to node k, and n_jk(i) signifies the number of shortest paths from node j to node k that passes through node i. The literature states that the distribution of food and beverage outlets in urban business districts is clustered, and the vitality of the food and beverage industry in the block may be negatively correlated with the distance to the nearest business district [45]. Therefore, we selected 13 main business districts within the study area (Figure 1) and calculated the straight-line distance from the block centroid to the nearest business district as a control variable. The distribution of urban restaurants exhibits a ring-shaped development pattern that decreases from the central urban area to the periphery [46,47]. Therefore, this study calculated the straight-line distance from the block centroid to the city center as a control variable, examining spatial location factors’ impact on the catering industry’s vitality.

3.2.4. Geographic Detector

Geographical detection involves a novel set of statistical methods for identifying spatial heterogeneity and its driving forces. The underlying concept is that a statistical correlation exists between two variables if their spatial distributions tend to be consistent. The assumption is that, if a certain independent variable significantly influences the dependent variable, their spatial distributions should exhibit similarity [48,49]. As there are few assumptions, geographic detectors can effectively solve problems, such as variable limitations in traditional statistics. Using factor detectors, we analyzed the extent to which the influencing factors accounted for the catering industry’s vitality. The value measures the impact of each influencing factor on the catering industry’s vitality, and is expressed as follows:

$$q=1-{{\displaystyle \sum }}_{h=1}^L{N}_h{\sigma }_h^2/N{\sigma }^2$$

in the formula, h represents the stratification of the influencing factor; $N_h$ and $N$ represent the number of units in layer h and the entire region, respectively; and ${\sigma }_h^2$ and ${\sigma }^2$ are the variances of the catering industry’s vitality in layer $h$ and the entire region, respectively.

Using an interaction detector, we probed the explanatory power of these two factors on the catering industry’s vitality under interactive effects. The geographic detector can identify not only the multiplicative interactions specified by econometrics, but also the strength, direction, linearity, and nonlinearity of the geographic detection factor interactions. In other words, the combined action of geographic detection factors X₁ and X₂ will enhance or diminish the explanatory capacity of the result factor Y. The evaluation method involves calculating the q-values of factors a and b for Y (q[X₁] and q[X₂]), calculating their interaction q-value (q[X₁∩X₂]), and comparing q(X₁), q(X₂), and q(X₁∩X₂). The relationships between these two factors can be divided into several types (

Table 3. Factor relationship in interaction detection.

Factor Relationship	Interaction
q(X1∩X2) < Min(q(X1),q(X2))	Nonlinear attenuation
Min(q(X1),q(X2)) < q(X1∩X2) < Max(q(X1),q(X2))	Single factor nonlinear attenuation
q(X1∩X2) > Min(q(X1),q(X2))	Double factor enhancement
q(X1∩X2) = q(X1) + q(X2)	Independence
q(X1∩X2) > q(X1) + q(X2)	Nonlinear enhancement

).

3.3. Data Description

The research data used in this study were primarily open source. The data source and basic situation are as follows: The restaurant data used in this study were obtained from Dianping.com (https://www.dianping.com, accessed on 1 June 2023) using Python web scraping techniques in June 2023. After extraction, the data underwent a series of processes including screening, cleaning, and coordinate correction to serve as the foundational data for the spatial analysis of the vitality of the catering industry. It is noteworthy that regarding data reliability, Dianping.com employs a review system that combines machine learning and human oversight to filter out obvious fraudulent activities and sensitive content. Therefore, the reviews and ratings displayed on the website possess a certain level of reliability. Additionally, the data were subjected to further cleaning processes such as deduplication and removal of outliers, ensuring that the overall data meet the reliability requirements of this study. Dianping, established in April 2003, serves as China’s premier urban consumption platform and an independent third-party consumer review website. The data cover a wide range of topics and are representative. After organizing the original data, 4935 units of restaurant data were obtained, including coordinates, merchant names, taste scores, service scores, atmosphere scores, and the number of comments. Block form data were obtained from the Beijing City Lab (https://www.beijingcitylab.com, accessed on 1 June 2023) and included the block building density, floor area ratio, functional mix index, and functional density. Their validity has been confirmed in previous studies [50]. The main city traffic road network data for Jinan were sourced from the China Geographic Information Public Service Platform (https://www.tianditu.gov.cn/, accessed on 1 June 2023). Other data were obtained from the Chinese Academy of Sciences Resource Environment Science and Data Centre (https://www.resdc.cn/, accessed on 1 June 2023).

4. Results

4.1. Spatial Visualization of Variables

Figure 2 shows that the distribution of vitality in the catering industry across the main urban area of Jinan is uneven; high vitality is observed in the south, east, and center, while the north, west, and outskirts exhibit lower vitality. The distribution of high-value areas in terms of the catering industry’s vitality closely matches that of business districts (Figure 1c). This indicates that the catering industry’s vitality is significantly influenced by the dominant function of the urban blocks. On the one hand, this phenomenon is a result of the profound geographical influence of the Yellow River and the southern mountainous area on the city’s form and functional zoning. On the other hand, it is caused by the agglomeration of catering service facilities in the main business districts, reflecting the spatial characteristics of social and economic elements flowing toward the main nodes in the market’s self-organization process.

Figure 3 shows the block form indicators and hotspot distributions in the study area. The hotspot analysis method is described in the literature [51]. The spatial distribution of block areas exhibits a clear hierarchical structure; the area size of the blocks in the central region is generally smaller, but gradually increases toward the periphery. This is related to the distribution of road networks in Jinan, where the central city area is divided into several smaller blocks by dense roads, while peripheral regions have sparser road networks, resulting in relatively larger blocks. The blocks in the western part of Jinan have a higher form ratio, suggesting a more regular plan form, whereas other areas have a more scattered distribution of high-form-ratio blocks. The average floor number of the blocks generally present east-high and west-low, and south-high and north-low spatial patterns. The blocks in the main eastern urban area have high land-use intensity and a higher average floor number. However, to protect the overall cityscape, Jinan strictly controls building heights in key areas within the cityscape belt. The building density in the blocks exhibit a north-high and south-low spatial distribution. Newly developed areas in the eastern part of the main urban area have adopted a high-rise, low-density development model that is conducive to creating open urban spaces, enhancing the image of the city and the attractiveness of the area. However, the low building density has resulted in more vacant land, which may lead to a lack of connectivity in height, facade, and functional use between plots; this may render the street interface discontinuous and incomplete. The high-value area of the functional density presents a southwest–northeast strip distribution, with the functional density gradient decreasing toward the periphery from this center.

4.2. Preliminary Exploration of the Relationship between Block Form and the Catering Industry’s Vitality

To explore the relationship between block form and the catering industry’s vitality, blocks were divided into high-, medium-, and low-vitality zones. Descriptive statistics were obtained from the block form indicators of different vitality zones to analyze the changing pattern of the catering industry’s vitality using different block form indicators. Specifically, blocks with vitality values lower than 0.06 were considered to be low-vitality zones (2017 blocks, 69.3%); blocks with vitality values between 0.06 and 1.00 were medium-vitality zones (625 blocks, 21.5%); and blocks with vitality values greater than 1.00 were high-vitality zones (269 blocks, 9.2%).

Figure 4 presents a comparison of the block form indicators for each vitality zone. From low- to high-vitality zones, the median, average, and standard deviation of the area indicators show a gradient increase, indicating that the catering industry’s vitality in blocks often increases with the increase in block area. The difference between the median and average of the form ratio indicator in different vitality zones is small, indicating that the change in the catering industry’s vitality in blocks with a form ratio is not significant. Furthermore, compared to the median and average of the average floor number indicator in low- and medium-vitality zones, those of the average floor number indicator in high-vitality zones show a slight increase, and the standard deviation lies between the two. This indicates that the average floor number of the block may be positively correlated with the catering industry’s vitality. Moreover, from low- to high-vitality zones, the median and average of the building density indicator both decrease, indicating that the building density may be negatively correlated with the catering industry’s vitality. For the functional density indicator, the median and average in high-vitality zones are the largest, and the degree of dispersion is the smallest, indicating that high-vitality zones often have higher functional density.

4.3. Spatial Heterogeneity Detection and Model Excellence Comparison

This study used the ordinary least squares (OLS) model for regression analysis. The results show that the variance inflation factor of each variable were all less than 1.825 (mean 1.281), indicating that there was no severe multicollinearity among the independent variables. Based on the OLS model, GWR4.0 software was used to test the geographic variability of variables, and the results are shown in

Table 4. Geographic heterogeneity test of variables.

Variable	F	DOF for F Test		DIFF of Criterion	Variable Type
Area	2.916	20.558	2542.828	−16.527	Local
Form ratio	2.036	20.649	2542.828	2.548	Global
Average floor number	1.644	18.834	2542.828	10.149	Global
Building density	7.971	19.951	2542.828	−119.980	Local
Functional density	0.800	20.353	2542.828	−48.698	Local
Population density	9.158	10.801	2542.828	79.402	Global
Bus stop density	3.374	13.549	2542.828	17.528	Global
Betweenness centrality	3.605	12.868	2542.828	19.803	Global
Distance to the nearest business districts	14.525	9.973	2542.828	−128.403	Local
Distance to the city center	14.578	6.392	2542.828	−83.583	Local

Note: DOF = Degrees of Freedom for F Test; DIFF = Difference in Criterion. If the DIFF of Criterion is positive, the variable does not exhibit geographical variability. If it is negative, the variable exhibits geographical variability.

. Variables such as area, building density, functional density, distance to the nearest business district, and distance to the city center show significant spatial non-stationarity and are local variables that should be subjected to local regression analysis. The form ratio, average floor number, population density, bus stop density, and betweenness centrality variables exhibit non-significant spatial non-stationarity and are global variables that should undergo global regression analysis. Therefore, MGWR 2.2 software was used to perform the regression to further analyze the local impact of block form on the catering industry’s vitality. From the results in

Table 5. Comparison of the GWR and MGWR models.

	OLS	GWR	MGWR
AICC	7508.891	7040.158	7006.384
Adjusted R2	0.094	0.339	0.420
Residual Sum of Squares	2476.450	1710.381	1580.663

Note: GWR = geographically weighted regression; MGWR = multi-scale geographically weighted regression.

, it can be seen that the goodness of fit of the MGWR model is 0.420, higher than that of the OLS and GWR models, and the AIC_C value (7006.384) is lower, indicating that the MGWR model is more suitable for this study. In addition, the small sum of squared residuals of the MGWR model indicates that its predicted values are closer to the true values.

4.4. Multi-Scale Geographically Weighted Regression Analysis

The statistical results of the MGWR model coefficients (

Table 6. Statistics of the regression results of the MGWR model.

Variable	Avg	SD	Min	Median	Max
Area	0.099	0.334	−0.178	−0.004	2.946
Form ratio	−0.010	0.002	−0.013	−0.010	−0.007
Average floor number	0.005	0.004	0.002	0.005	0.012
Building density	0.076	0.241	−0.297	0.011	1.992
Functional density	0.045	0.144	−0.589	0.069	0.352
Population density	0.131	0.006	0.119	0.133	0.139
Bus stop density	0.092	0.066	0.019	0.068	0.270
Betweenness centrality	0.123	0.003	0.116	0.124	0.126
Distance to the nearest business districts	−0.591	0.944	−2.794	−0.141	0.497
Distance to the city center	0.643	0.857	−1.063	0.730	2.213

Note: SD = standard deviation.

) indicate that the coefficients of average floor number, population density, bus stop density, and betweenness centrality are positive, suggesting that these four factors promote the catering industry’s vitality. Contrarily, the coefficient of form ratio is negative, implying that regular block forms hinder the catering industry’s vitality. Moreover, other variables have different effects on the catering industry’s vitality in different locations. The standard deviation can indicate the degree of dispersion of the impacts of various variables on the catering industry’s vitality; the larger the standard deviation, the greater the difference in the impact of different regional factors on the catering industry’s vitality. Specifically, area, building density, functional density, distance to the nearest business districts, and distance to the city center have significant spatial differences in their impact on the catering industry’s vitality, which validates the results of the geographical variability test of the variables.

A thorough analysis of the spatial heterogeneity of local variables (Figure 5) yields the following results: Block area predominantly contributes positively to the catering industry’s vitality and presents a “center–periphery” spatial differentiation characteristic. As proximity to the city center increases, so does the positive impact of the increase in block size on the catering industry’s vitality. Among all the block form factors, the spatial difference in the regression coefficient of area is the largest. The regression coefficient of most areas is less than 1; however, the regression coefficient of the city center is between 1.940 and 2.946, indicating that the block area factor is the primary element affecting the catering industry’s vitality in the city center. The positive effect of building density on the catering industry’s vitality presents a distinct circular structure. The regression coefficient gradually decreases towards the periphery, focusing on Spring City Square. This indicates that the promoting effect of an increase in block building density on the catering industry’s vitality decreases as the distance to the city center increases. The effect of functional density on the catering industry’s vitality is more complex. The regression coefficient on the west side of Spring City Square is negative, indicating that crowded commercial facilities and the noisy environment they create have a negative impact on the dining experience. Contrarily, the area east of Spring City Square balances the agglomeration of commercial facilities.

4.5. Detecting the Influencing Factors in Different Dominant Functional Blocks

Figure 6 shows that the land-use types in the city’s main urban areas have a concentric circular layered structure. From the city center to the outside, there are commercial function, residential function, mixed function, public service function, and industrial and logistics blocks. Other lands, mainly green spaces and parks, are distributed throughout all the layers. The spatial distribution of blocks with different dominant functions reflects the objective economic laws followed by urban land-use. To prevent the dominant function of the block from impacting the catering industry’s vitality, the explanatory degree and its interaction with the influencing factors in the residential, public service, commercial, and mixed functional blocks were detected. The single driving factor results show that in all detection results of the blocks, the q-values of the detection factors (0.002–0.076) are relatively small, and the impact of each factor on the catering industry’s vitality is equivalent, without any significant decisive factors (

Table 7. Detection results of influence factors of all and different dominant functional blocks.

	All	Residential	Public Service	Commercial	Mixed
Area	0.002	0.010	0.025	0.024 *	0.032
Form ratio	0.005 **	0.012	0.026	0.031 **	0.030
Average floor number	0.002	0.015	0.022	0.027 *	0.006
Building density	0.011 ***	0.065 ***	0.007	0.047 ***	0.242 ***
Functional density	0.067 ***	0.053 ***	0.064 ***	0.066 ***	0.095 ***
Population density	0.027 ***	0.028 ***	0.024	0.084 ***	0.107 ***
Bus stop density	0.037 ***	0.018 *	0.083 ***	0.065 ***	0.027 *
Betweenness centrality	0.050 **	0.051 ***	0.017	0.090 ***	0.127 ***
Distance to the nearest business districts	0.076 ***	0.089 ***	0.024	0.160 ***	0.117 ***
Distance to the city center	0.076 ***	0.099 ***	0.015	0.178 ***	0.185 ***

Notes: * p < 0.1, ** p < 0.05, *** p < 0.01.

).

Separately determining the factors for blocks with different dominant functions highlighted the decisive role of the specific detection factors. From the perspective of different dominant functions, the influence of block form on the catering industry’s vitality in commercial functional blocks is the most obvious, while its influence on public service functional blocks is the weakest. Moreover, among the block form indicators, building and functional densities have a significant impact on the catering industry’s vitality in the residential and mixed functional blocks. Form ratio, building density, and functional density play the greatest roles in enhancing the catering industry’s vitality in commercial functional blocks. Functional density is the only block form element that has a significant impact on the catering industry’s vitality in the public service functional blocks.

From the perspective of detection factors, the changes in the influence of each factor on the catering industry’s vitality in blocks with different dominant functions show specific consistencies and differences. The determining powers of the functional and bus stop densities in blocks with different dominant functions are relatively consistent. Large differences exist in the impacts of form ratio, building density, population density, and betweenness centrality on the catering industry’s vitality in blocks with different dominant functions, whereas other factors show a smaller degree of change in their influence capacity.

The interaction detection results show that the interactions between pairs of factors mainly manifest as dual-factor enhancement and nonlinear enhancement; that is, the impact of dual-factor interaction surpasses that of individual factors. The catering industry’s vitality stems from the combined effect of various factors. The interactive driving factors that significantly influence the catering industry’s vitality in residential functional, public service functional, commercial functional, and mixed functional blocks are average floor number ∩ distance to the city center (0.276), area ∩ bus stop density (0.379), bus stop density ∩ distance to the city center (0.324), and betweenness centrality ∩ distance to city center (0.528), respectively (Figure 7). From the perspective of detection factors, the q-values of the four driving factors—building density, functional density, distance to the nearest business districts, and distance to the city center—are generally higher than those of the other single driving factors and interactive driving factors, which is consistent with the results of the spatial regression analysis. Population density has a significant impact on single driving factors, while area, form ratio, average floor number, betweenness centrality, and bus stop density are the five driving factors that are significant for interactive driving factors.

5. Discussion

5.1. Comprehensive Impact Mechanism of Block Form on the Catering Industry’s Vitality

We found that the effect of block form on the catering industry’s vitality is homogeneous and heterogeneous, depending on different locations and dominant functional blocks.

Without considering the dominant function of the block, our analysis shows that form ratio and average floor number have a consistent impact on the catering industry’s vitality, overall. In other words, blocks with vibrant catering facilities usually have irregular planar morphologies and high-rise buildings. This may be related to the fact that people are increasingly paying attention to experiential consumption [52], focusing on the atmosphere, mood, and feelings during dining, and not just the product itself. The combination of irregular block forms and high-rise building groups not only maximizes the integration effect of space, but also increases the richness of people’s spatial experiences, making the interaction between people and dining spaces more intense and of higher quality. Furthermore, the impacts of area, building density, and functional density on the catering industry’s vitality show obvious spatial differences. Proximity to the city center, larger area, and higher building density offer the advantages of the central location in terms of population, transportation, and so on, which impact the catering industry’s vitality positively and significantly. This shows that although the widespread use of mobile Internet and information technology in the catering industry has a profound impact on the distribution pattern of dining spaces [53], the central place theory is still applicable to the study of dining spaces in the mobile information era. The effect of block functional density on the catering industry’s vitality has no obvious pattern, indicating that the functional density index has a strong complex comprehensiveness with other influencing factors.

Considering the actual situation of the dominant function of the blocks, our research found that, among the morphological factors of residential functional blocks, building and functional densities have a key impact on the catering industry’s vitality. Residential function is the basic function of a city as a human living space, and the blocks dominated by the residential function have a dense urban population. Simultaneously, residential buildings are not only residential functional blocks, but they also require commercial, cultural, entertainment, greening, and other grassroots public service facilities for the residents. Blocks with high building density have a higher population; meanwhile, the development mode of high building density can reduce the distance between buildings on adjacent plots, improve the interface on both sides of the street, and provide favorable conditions for restaurant location selection. Higher functional density attracts people from this and other blocks, which means more customers for restaurants. The catering industry’s vitality in residential functional blocks with a high average floor number close to the city center is relatively high, indicating that increasing land-use intensity can fully utilize the locational advantages of the city center area, which has a positive impact on creating vigorous catering industry vitality.

The influence of the block plan and architectural forms on public service functional blocks is small, and functional density is the only morphological index that can significantly influence the catering industry’s vitality. Specifically, bus stop density has a larger q-value in both single driving factor detection and interaction driving factor detection, resulting in public service functional blocks. Public service functional blocks mainly include government departments and state-owned enterprises with high commuting needs, and are prone to traffic congestion. Therefore, convenient public transportation has an obvious promotional effect on the catering industry’s vitality.

The catering industry’s vitality in commercial functional blocks is most influenced by block form. The area and average floor number can significantly promote the catering industry’s vitality by increasing the population capacity of the blocks, and thereby the demand for catering. The boundary space of the block is the area with the most active crowd and the highest frequency of interaction between residents and the catering space. Irregularly shaped blocks often have a larger front shop area, resulting in more public parking lots and walking areas. As Jacobs said, for a vibrant block, “sidewalks must be wide enough for children to play freely in it; streets must be short, so that pedestrians always have a fresh feeling of turning” [3]. Similar to other dominant functional blocks, blocks with high functional density promote the catering industry’s vitality by increasing the attractiveness for the active population. The market size in the city center is large, and the flow of people and vehicles is dense; therefore, it needs to be equipped with perfect public transportation facilities to maximize locational advantages.

For mixed functional blocks, blocks with higher functional density and building density can effectively provide activity venues and communication platforms for residents, thereby enhancing the catering industry’s vitality. It should be noted that, compared with other dominant functional blocks, the combined interaction of morphological and socioeconomic elements of mixed functional blocks has a more obvious promoting effect on the catering industry’s vitality. Specifically, the betweenness centrality of the road network has a larger q-value in combination with other factors. This indicates that the use of the block open space system to connect urban elements, integrating urban functions and landscape systems into the block open space, has important significance for enhancing the catering industry’s vitality.

5.2. Block Planning Inspiration for the Enhancement of the Catering Industry’s Vitality

The block serves as a crucial element in the urban built environment, and constitutes the fundamental unit for urban planning and management. Building high-quality urban forms plays a significant role in improving urban land-use efficiency and enhancing urban spatial vitality. From the perspective of enhancing the catering industry’s vitality, considering the organizational logic between the morphological elements of the blocks and the residents’ activity patterns, we propose the following specific recommendations. First, we develop irregular block forms to form shared blocks with the city. Irregular block plan forms can enhance the catering industry’s vitality, especially commercial functional blocks. Therefore, developing branch-shaped, radial, or free-style block forms, according to local conditions, is conducive to creating a vigorous catering industry; however, we should avoid negatively impacting urban main road traffic. Open blocks have the advantage of creating diversified urban styles, relieving urban traffic pressure, and establishing vibrant public streets and lane spaces. Integrating blocks with urban road network structures, public service facilities, and landscape systems is essential for sustainable urban development. Second, the building and functional densities should be appropriately increased to promote efficient block operations. The spatial agglomeration of rich functions is a catalyst for creating blocks with spatial vitality [54]. The goal is to provide for and meet the daily needs of residents on a smaller spatial scale, ultimately improving land utilization and compactness. During the urban renewal and redevelopment process, we should effectively integrate various urban spatial functions such as production, living, and business services, to create a highly integrated vibrant open space.

6. Conclusions

The main findings of this study can be summarized as follows: irregular block plan forms and high-rise buildings promote the catering industry’s vitality; the direction of the effect of block area, building density, and functional density on the catering industry’s vitality has spatial heterogeneity; and indices such as betweenness centrality, distance to the nearest business districts and distance to the city center have a significant impact on the catering industry’s vitality. Specifically, block form has a significant impact on blocks with different dominant functions. Block form factors have the most obvious impact on the catering industry’s vitality in commercial functional blocks and the weakest impact in public service functional blocks. Building and functional densities have a significant impact on the catering industry’s vitality in residential and mixed functional blocks. Functional density is the only block form element that has a significant impact on the catering industry’s vitality in public service functional blocks. Finally, the superior block form can, to some extent, reduce the dependence of the catering industry’s vitality in this block on the regional population and public transportation facilities; however, a convenient public transportation system and network node positions can play a connecting role can better exert the advantages of block form.

This study’s findings carry policy implications in terms of the pivotal role played by block forms in creating vibrant cities. It is therefore necessary to plan urban areas scientifically for different regional socioeconomic conditions. The government should make full use of the relationship between block form elements and socioeconomic factors such as population, traffic, and location to thereby ensure that they complement each other. Urban designers and planners should recognize that with the already determined block form, they can use the different requirements of blocks with different dominant functions for spatial morphology elements to reasonably allocate commercial facilities for the catering industry or other types of economic activities.

Despite its clear contributions, this study has some limitations. First, social network review data are limited in terms of representativeness and universality. As we used online review data to quantify the urban catering industry’s vitality, the results may deviate from reality. Second, when establishing a system of factors influencing the catering industry’s vitality in block form, this study mainly considered location and traffic factors; however, the catering industry is also affected by local political, economic, cultural, and time-related uncertainties. Lastly, due to data availability, this study only selected a portion of representative block form indicators, inevitably omitting some potentially valuable block form factors. Based on these considerations, future research should use more accurate and larger databases and more comprehensive evaluation standards and quantification systems to provide more beneficial references for the creation of urban vitality and block planning and construction.