*2.3. Methods*

### 2.3.1. Evaluation of Rural Development Potential

1 Establishment of an index system

Rural development potential is a comprehensive reflection of the interaction of many factors, such as resource endowment, functional positioning, location conditions, development policies, and the historical background of each region [29]. Rural resources and their ability to use resources are of grea<sup>t</sup> significance to their development and revitalization [30,31]. Current studies on rural development potential are mostly focused on the potential for intensive rural land use [32], rural tourism development [33], and rural settlement improvement [34,35]. For the selection of indicators, this study extensively referred to the research literature related to rural evaluation. Zhang R.T. et al. selected three indicators of population development, industrial development and land use to construct the evaluation system of rural development level [36]. In the study of comprehensive evaluation and classification of rural development, Han X.Y. et al. selected seven categories of elements for evaluation, including agricultural production, non-farm economy,

daily life, social management, facility services, natural conditions, and human resources, according to three major functions: living, production, and ecology [37]. In evaluating the characteristics of rural transformation, Long H.L. et al. constructed an evaluation system from three dimensions: rural economic development, agricultural production development, and rural social development [38]. The core elements of rural development potential evaluation through literature research should include location, resource conditions, and socio-economic development status. Rural development potential represents the ability to attract people, capital, and technology in the future. Therefore, the attraction of internal and external powers constitutes the main support of its development potential. The internal power is determined by the location of the village and its resource endowment. By contrast, external power refers to the support of external resources for rural development. This notion indicates that further opportunities for development resource elements can be obtained to accelerate the contribution of external resources to its construction, reflecting the industrial advantages of rural development and its ability to attract social capital. In addition, labor and productivity levels, including livelihood and production security and ecological security, pose obstacles to rural development. Such constraints are key factors affecting rural development potential.

The common indicators applicable to this study were sorted out by combing the relevant literature on village evaluation with high-frequency indicator screening. Then, through field research, individual indicators were screened out considering the objectives and characteristics of village development in Shapingba District. Finally, the evaluation index system of rural development potential including four dimensions and 18 specific indicators is established (Table 2). The indicators comprehensively reflect the rural development potential and conform to the principles of scientific method and operability. The indicators of location advantage measure the accessibility of villages to the outside world from the aspects of the average elevation, topographic relief, distance to the nearest township, and road network density. Resource endowment includes the connotation of natural and human resources. Thus, the indicators of per capita construction land area, per capita cultivated land area, per capita garden land area, the number of natural and cultural landscape resources, and the percentage of communist party members in the village are selected for characterization. Economic vitality is an important factor in attracting social capital and technology. We select indicators, such as the proportion of the output value of three types of industries, the number of township enterprises and family workshops, and the level of agricultural mechanization to characterize the development of village industrial structure and productivity level. The aging level of the population, the proportion of migrant workers in the total population, the proportion of the area of the high-prone region of geological disaster, and the proportion of redlines for protecting the area of the ecosystem in village area were selected to represent the degree of construction in village development. The consistency reliability (Cronbach's alpha) value of the rural development potential measurement index system established was tested to be 0.786, which is greater than the empirical threshold of 0.7 and meets the requirements of index representation and consistency.



2 Data standardization

Range standardization is a method to standardize positive and negative indicators in economic statistical analysis, which is a linear transformation of original data. The range is obtained by calculating the difference between the maximum and minimum values of the index. All index values are mapped to [0, 1]. The calculation formulas are as follows:

When Xij is a positive indicator,

$$Z\_{\rm ij} = \frac{\chi\_{\rm ii} - \min \chi\_{\rm iii}}{\max \chi\_{\rm ii} - \min \chi\_{\rm ij}},\tag{1}$$

when Xij is a negative indicator,

$$Z\_{\text{i}\text{j}} = \frac{\max \chi\_{\text{i}\text{j}} - \chi\_{\text{i}\text{j}}}{\max \chi\_{\text{i}\text{j}} - \min \chi\_{\text{i}\text{j}}},\tag{2}$$

where Zij denotes the standard values for raw data, and Xij denotes the specific index value of a sub-item.

3 Indicator weight set

The entropy weight method is an objective weighting method, which overcomes the subjectivity and randomness brought by the subjective weighting method. The application of the entropy weight method can make the evaluation result more in accord with the actual situation. To avoid the disadvantage of the insufficient scientific meaning of the weight results brought by objective assignment, we use Analytic Hierarchy Process to revise the weight results and comprehensively determine the weight. The steps are as follows:

First, the translation of dimensionless data is coordinated,

$$X\prime\_{\rm ij} = Z\_{\rm ij} + \mathbb{C}\_{\prime} \tag{3}$$

where *X'ij* is the index value after data standardization and translation, and *C* is the translation amplitude (in this study, *C* = 0.0001).

Second, the information entropy of the index is determined,

$$\mathbf{e}\_{\rangle} = -\left(\frac{1}{\ln \mathbf{n}}\right) \times \sum\_{i=1}^{\mathbf{n}} Z\_{\mathbf{i}\mathbf{j}} / \sum\_{i=1}^{\mathbf{n}} Z\_{\mathbf{i}\mathbf{j}} \ln \left( Z\_{\mathbf{i}\mathbf{j}} / \sum\_{i=1}^{\mathbf{n}} Z\_{\mathbf{i}\mathbf{j}} \right),\tag{4}$$

where ej is the information entropy of the *j*-th evaluation index, and n is the number of evaluation units (n = 38 in this study).

Third, the index weight is determined,

$$Q\_{\rangle} = 1 - \varepsilon\_{\rangle} / \sum\_{j=1}^{m} (1 - \varepsilon\_{j}) , \tag{5}$$

where *Qj* is the weight of the *j*-th evaluation index, and *m* is the number of evaluation indexes (*m* = 18 in this study).

Finally, according to analytic hierarchy process, all indicators are divided into groups according to the correlation and affiliation. Each group is defined as a layer, and finally, a hierarchical system structure model associated with a combination of the highest, middle, and lowest layers is created (Figure 2). The consistent matrix method is used to compare the evaluation factors of the same layer from the second layer to compare their importance relative to the previous layer. Generally, the 1~9-bit scale method is used to construct the judgment matrix. The maximum characteristic roots and the corresponding eigenvectors of the above judgment matrix are calculated, and then the consistency of the matrix is tested using the consistency index, the average random consistency index and the consistency ratio. Referring to existing studies and consulting with experts, we constructed judgment matrices and past consistency tests [39]. The above steps are realized by YAAHP software to realize the calculation process. Then we obtain the weight Fi of each index. The indicator weights were revised using the preference coefficient μ to derive the final weight.

$$\mathcal{W}\_{\dot{l}} = \mu F\_{\dot{l}} + (1 - \mu) Q\_{\dot{l}\nu} \tag{6}$$

where <sup>W</sup>*j* is the final weight of the *j*-th evaluation index, and *Fi* is the weight calculated by Analytic Hierarchy Process. μ is the preference coefficient (μ = 0.5 in this study).

**Figure 2.** Hierarchical structural mode.

> 4 Evaluation model of rural development potential

Combined with the standardized value and weight of each evaluation index, the rural development potential in each evaluation unit is calculated. The calculation formula is as follows:

$$D\_i = \sum\_{j=1}^{m} \mathcal{W}\_{\dot{j}} Z\_{\dot{i}j\prime} \tag{7}$$

where *Di* is the score of the rural development potential.

### 2.3.2. Hierarchical Cluster

Cluster analysis is a method of classifying samples. The basic principle is to determine quantitatively the relationship between samples according to their attributes. The hierarchical cluster method is the most widely used clustering analysis at home and abroad. This method first regards the clustered samples or variables as a group, then determines the similarity statistics between classes, selects the closest two or several classes to merge into a new class, and calculates the similarity statistics between the new class and other classes. Then the closest two or several groups are selected to merge into a new class until all samples or variables are merged into one class. In this paper, the 38 administrative villages are taken as the basic units, and the research units are partitioned by hierarchical clustering. The specific steps of cluster analysis are as follows:

First, the Euclidean distance is selected to define the distance between samples:

$$d\_{\rm ij} = \sqrt{\sum\_{k=1}^{m} \left(X\_{ik} - X\_{jk}\right)^2} \tag{8}$$

where *d*ij is the distance between samples *i* and *j*, m presents the number of dimensions, and *Xik* and *Xjk* are the evaluation values of samples *i* and *j* on the *k*-th dimension, respectively.

Then, the distance coefficient *d*ij between any two sample points can be calculated in turn to obtain a distance matrix between samples:

$$\mathbf{D} = \begin{pmatrix} d\_{\mathrm{i}\mathbf{j}} \end{pmatrix} = \begin{bmatrix} d\_{11} \ d\_{12} \ \ldots \ \ldots \ \ldots \ d\_{1n} \\ d\_{21} \ d\_{22} \ \ldots \ \ldots \ \ldots \ d\_{2n} \\ \ldots \ \ldots \ \ldots \ \ldots \ \ldots \\ d\_{n1} \ d\_{n2} \ \ldots \ \ldots \ \ldots \ d\_{nn} \end{bmatrix} . \tag{9}$$

Finally, the longest distance method is used for clustering. If *Xi* is any sample in class Gp and *Xj* is any sample in class *Gq*, then the longest distance in classes *Gp* and *Gq* is as follows:

$$\mathbf{D}\_{p\eta} = \max\_{\mathbf{X}\_i \in \mathbf{G}\_{p\prime}, \mathbf{X}\_j \in \mathbf{G}\_q} d\_{ij\prime} \tag{10}$$

among them, the smaller the <sup>D</sup>*pq* is, the smaller the distance between samples is. The closer the properties of samples i and j are, the more they can be divided into the same type.

### **3. Result and Analysis**

*3.1. Analysis of Rural Development Potential*

3.1.1. Spatial Distribution Characteristics of Rural Development Potential

To clearly demonstrate the spatial distribution characteristics of rural development potential, the comprehensive scores obtained in Equation (7) were used. With Arcgis 10.2, the scores were spatially linked with each research unit in the form of vector data and partitioned using natural breaks [40]. We divided them into high, medium, and low levels and plotted the spatial distribution (Figure 3).

**Figure 3.** Spatial distribution of rural development potential in Shapingba District.

According to the calculation results, the average, minimum and maximum values of the rural comprehensive development potential are 0.2263, 0.3312, and 0.4602, and the proportion of the number of administrative villages in high, medium, and low levels is 13.16%, 73.68%, and 13.16%, respectively. Villages with medium development potential have a clear numerical advantage. The spatial distribution pattern of the rural development potential is centered on the central and southern urban development areas, gradually decreasing toward the peripheral area. The villages with a high potential value of rural development are mainly distributed in the south and central regions, whereas those with a low score are mainly concentrated in the northern fringe region. The overall development potential tends to be balanced, but the advantages and development obstacles of villages in different regions vary. The southern villages are close to the regional development center, with evident location advantages, a high degree of urban–rural development integration, and greater economic development vitality, which largely compensate for the disadvantage of the shortage of background resources and thus present good development potential. The northern center of the Shapingba District is the second core area of its urban development. With the strengthening of land acquisition, the urban area has been expanded, and the advantage and obstacles of village development at its edge are evident. In terms of development advantage, the villages in the region are rich in land resources and have a high number of natural and humanistic landscapes. The obstacle to development mainly comes from the strict control of redlines for protecting the ecosystems, making it difficult to meet the basic demand for construction land. The villages in the northern periphery are far away from the urban centers. The infrastructure construction is relatively backward, and the large influx of rural population into the cities has led to the phenomenon of the hollowing out of villages.

### 3.1.2. Characteristics Analysis on the Sub-Dimension

To better portray the spatial pattern of rural development potential in each subdimension and its divergent characteristics, we used the same method to divide the scores of each dimension into high, medium, and low levels and plotted the spatial distribution of the sub-dimension (Figure 4).

**Figure 4.** Distribution of rural development potential.

Location advantage (Figure 4a): On the whole, the location pattern of all villages is good, the average score of location advantage is 0.0967, and the proportion of the number of administrative villages in high, middle, and low levels is 39.47%, 42.11%, and 18.42%, respectively. The spatial distribution shows the trend of taking the north center as the core and gradually decreasing around. High-level areas are concentrated in the suburbs of the two town development centers in Shapingba District. The reason contains two aspects, one is that the area receives strong radiation from the urban transportation network, has good road facilities, and has good external connection conditions; the other is that 69% of the villages in the high-level area have a gentle average elevation and topographic relief, making village construction less difficult, which is also a key reason for its good performance in terms of coordination of location pattern. The medium-level villages are located on the periphery of the high-level ones, with the largest number of villages and slightly poorer external connections. The low-level areas are mainly located at the edge of Shapingba District, mainly because of the weak external communication due to the underdeveloped transportation network.

Resource endowment (Figure 4b): Its average score was 0.0528, with significant village differences, and the proportion of the number of administrative villages in high, medium, and low levels is 13.16%, 36.84%, and 50.00%, respectively. Spatially, the distribution trend of "high value in the north and low value in the south" is observed. The high-level areas are scattered in a dotted pattern in the south of the district. These villages have abundant land resources and good resources and conditions for tourism development. The medium-level villages are mainly distributed in a contiguous manner in the northern part of the region. The villages in the low-level areas are mainly located in the southern district. Considering their proximity to the town centers, land urbanization is developing rapidly. The loss of cultivated and garden land is serious, and the agricultural production capacity is low.

Economic vitality (Figure 4c): Its average score was 0.0351, and the proportion of the number of administrative villages in high, medium, and low levels is 15.78%, 42.11%, and 42.11%, respectively. The spatial distribution shows a trend of decreasing from south to north. All of the high-level areas are located in the southern part of Shapingba District, where rural enterprises are well developed, and the location advantage of being close to the urban makes them closely connected to the needs of urban residents and the development of urban industries. The industrial structure is dominated by the secondary and tertiary industries. The medium-level areas are mainly distributed in the northern part of the region. The villages in the low-level area are mainly located in the southern fringe area, where the poor foundation of industrial development and low productivity and organization levels are important development obstacles in the area.

Development constraints (Figure 4d): The higher the score of this dimension, the smaller the resistance to the spatial development of the village. Its average score was 0.1466, and the proportion of the number of administrative villages in high, medium, and low levels is 34.21%, 44.74%, and 21.05%, respectively. The spatial distribution is most widely distributed in the high and medium-level areas. The two main reasons are as follows: (1) population loss is relatively small; and (2) ecological control is moderate, and relatively free space is available for development.

### *3.2. Classification of Village Types*

The spatial clustering function of SPSS Statistics 26 software was used to realize the calculation process of Equations (8)–(10) and output the system clustering results. Based on the clustering results, a total of 38 administrative villages in the district were divided into four types, namely, core planning area, important planning area, general planning area, and basic control area. The spatial visualization results were realized using ArcGIS 10.2 (Figure 5).

(1) Core planning area: The village type has five administrative villages, that is, Sanhe, Renhemen, Yangjiamiao, Xinfa, and Yongningsi. These villages have excellent geographical locations, mostly located in suburban integration areas or township centers.

They have evident advantages in the four sub-dimensions and have grea<sup>t</sup> potential for comprehensive rural development. They should have priority in the rural revitalization strategy in order to develop as a regional growth hub. They must also lead the surrounding villages to develop together. The villages in the core planning area should be given the highest level of attention and detail in the preparation of practical village planning to support of their scientific and orderly development.


**Figure 5.** Results of village type.

### **4. Differentiation Strategies for Practical Village Preparation**

According to the positioning of villages and the actual needs of national space development and protection, the level of the development potential of villages should be considered, and the preparation requirement of practical village planning must be reasonably determined. Combined with the requirements of the latest technical specifications for the preparation of practical village planning in Chongqing, the planning requirements should include nine items: development orientation and objectives, territorial space control and layout, industrial development layout, rural residential area planning, infrastructure and basic public service facilities layout, ecological protection and land consolidation, historical and cultural preservation and heritage planning, rural style guide, safety and disaster prevention and mitigation planning (Table 3). However, these requirements are not mandatory for all village plans, and each village can choose the necessary contents and expanded contents in a scientific and reasonable way according to its development type and need, where the former is mandatory for the village plan, whereas the latter is selected in conjunction with the actual needs of the village. According to the basic requirements of the technical specifications, the first two planning requirements shall be necessary for all villages, and the rest shall be selected according to different types of villages. This paper classifies the villages into four types: core planning area, important planning area, general planning area, and basic control area, based on the level of rural development potential. With reference to the current content selection of village planning in Chongqing, this paper identifies the necessary and expanded planning content for different types of villages and proposes corresponding planning strategies based on the authors' field project experience. The level of detail required by the four types of village planning varies, and the corresponding plan content should also make a difference. To clearly demonstrate the study, we selected one case village in each of the four types for illustration. The case villages were Sanhe, Qingfengshan, Shiyuan, and Maoshanxia (Figure 6). The village plans shown in Figure 6 are all projects undertaken by our research team in 2019.

### *4.1. Core Planning Area*

The core planning area is in the urban–rural transition zone, receiving stronger radiation from urban resources, and generally has the advantages and potential to become a back garden of the city. This area also has the conditions to transform into a city. To a certain extent, the core planning area has the ability to serve urban development, undertake urban function spillover, and meet urban consumption demand [41,42]. This area is the front-runner of urban–rural integration development. In terms of the level of detail, the core planning area has the highest requirements for the content of the practical village plan, which should meet the nine requirements. In terms of planning strategies, we use the land layout planning of Sanhe Village as an example (Figure 6a). First, priority is given to guaranteeing the demand for land for the construction of public service facilities, such as education, culture, and medical care, improving their construction level and service quality, forming a public service network with villages in the core planning area as the central nodes, and improving the attractiveness of the countryside. Second, the construction of a railway network is focused to deepen regional connections, to attract external resources into the countryside, and to help rural revitalization. Third, this planning also focuses on improving the efficiency of rural construction land use, effectively developing unused land, and stimulating rural development while safeguarding land for the development of rural advantageous industries.


**Table 3.** Contents of practical village planning.

Note: Necessary content - Expanded content.

### *4.2. Important Planning Area*

The important planning area is the main area to support district development and the area to guarantee ecological security. The latest functional positioning of Chongqing has positioned the rural area of Shapingba District as the "Western International Slow City" and the "Beautiful Back Garden for Citizens' Leisure". Vigorously developing modern urban agriculture in urban suburbs will become the main direction of agricultural and rural reform in the Shapingba District and the important direction of its village planning. The village planning of the important planning area includes seven necessary contents. In terms of planning strategies, we take the village of Qingfengshan Village as an example (Figure 6b). First, we can plan and implement high-quality projects for leisure agriculture and rural tourism and build a number of leisure and tourism parks with complete facilities and diverse functions, forest homes, recreation bases, rural bed and breakfast, and small towns with special features. Second, to form a development model integrating humanities and ecology into cultural tourism, we consider the following: inheriting local culture, identifying the core cultural elements of rural regional space, discovering cultural tourism resources with uniqueness and attractiveness, using traditional villages, scientific and technological agriculture, idyllic scenery, agricultural production landscape, green ecological resources, and other elements. Third, we aim to deeply improve road networks, water and drainage networks, power grids, and communication networks, including other infrastructure construction to provide a good development environment for industries in the district.

**Figure 6.** Land use layout planning for different types of villages.

### *4.3. General Planning Area*

The general planning area is a key area for ecological protection, and the necessary content of its village planning should contain four basic contents, namely, development orientation and objectives, national land control and layout, industrial development layout, and ecological protection and land consolidation. In terms of planning strategies, we take the village of Shiyuan Village as an example (Figure 6c). First, we strengthen the protection

of ecological land, guide the gradual withdrawal of environment-consuming industries, and develop ecological industries. Second, we focus on ecological safety restoration and reduce the adverse impact of geological disasters on the production and life of villagers. Third, a certain percentage of construction land is reserved for mobile indicators to support scattered rural cultural and tourism facilities and new rural industries.

### *4.4. Basic Control Area*

The background conditions and economic development environment of the basic control area are relatively backward, and its village planning requires the lowest level of detail. The necessary content only needs to cover two items, that is, development orientation and objectives, and national land control and layout. Taking Maoshanxia Village as an example (Figure 6d), in terms of planning strategy, the first point is to strictly implement spatial use control and strictly control the phenomenon of illegal land use. The second one is to inherit vernacular culture, including vernacular landscapes, humanistic and historical relics, and traditional farming implements, and to protect traditional vernacular architecture and prevent major demolition and construction. Finally, efforts should be made to enhance the infrastructure construction in farmers' clusters, improve the efficiency of public services, and meet the basic needs of residents for water, electricity, transportation, and communication.
