Sustainability Constraints on Rural Road Infrastructure

Abstract

Research on the sustainability of rural roads is of great significance to the integrated promotion of rural habitat improvement, the maintenance of regional ecological patterns, and the implementation of the rural revitalization strategy. This paper examines the constraints to ensuring the sustainability of road infrastructure in rural Shaanxi, China. Rural road infrastructure plays an important role in poverty alleviation. Rural Shaanxi Province is situated among a variety of terrains and spans a large area. Therefore, using the current situation of rural roads in the province as a starting point, the constraints and their rankings that impede the sustainability of road infrastructure are identified through the fuzzy Delphi method (FDM) and structural equation modeling (SEM), which in turn suggests governance measures that can be taken. The data in this paper come from 498 valid questionnaires obtained from 50 townships and 101 sample villages in Shaanxi Province. Due to the huge number of questionnaires, the questionnaires were tested with the help of SPSS 26.0 software, which showed that the questionnaires had high reliability and validity, and then the structural equation model was validated and modified according to the recommendations of goodness-of-fit statistics measurement and the modification index. Finally, the key factors affecting the sustainable development of rural road infrastructure were finally identified.

1. Introduction

The goal of sustainable development [1], which is often aimed for in the construction industry for green buildings [2], it has been gradually applied to road infrastructure in recent years [3]. Rural road infrastructure plays an important role in poverty alleviation and contributes to the socio-economic development of rural communities [4]. Therefore, this study extends sustainability research to rural road infrastructure development and emphasizes environmental, economic, and social balance. Rural road infrastructure sustainability is subject to a number of constraints [5], which have been neglected in previous studies. Without a proper understanding of these constraints, it is difficult to develop and take effective measures to promote sustainable road infrastructure development in towns and cities. Shaanxi Province in China has a large number of rural areas [6], contains a variety of topographic landscapes, and spans a wide range of areas. The southern Shaanxi region [7] is located in the southwest of Shaanxi Province, which is bounded by the Qinling Mountains in the north, the Bashan Mountains in the south, and the Basin in the center; the Guanzhong Plain extends from Baoji in the west to Tongguan in the east, with an average elevation of 520 m; the northern Shaanxi region includes the cities of Yulin and Yan’an, which are located in the Loess Plateau’s hilly, gully area; and Yulin City is roughly bounded by the Ancient Great Wall, with a windy, sandy, grassy beach area in the north [8], and the Loess hills and gullies area in the south [9]. Therefore, studying rural road sustainability in Shaanxi Province will provide a basis for determining the constraints to rural road infrastructure sustainability [10] in other regions. The significant differences in natural conditions, economic foundations, and social environments in different areas of Shaanxi Province pose unique challenges to the sustainable development of rural road infrastructure. In view of this, it is particularly important to explore in depth the sustainable development trajectory of rural roads in conjunction with domestic and international infrastructure sustainability assessment methods.

In the past decade, the results of research on China’s new rural construction [11] and infrastructure have been remarkable. Sierra and other researchers have introduced a variety of methods for assessing the sustainability of infrastructure [12]. Earlier in 2013, through a questionnaire survey completed by experts from different organizations and institutions, the importance indices of project evaluation indicators were statistically analyzed, and the correlation coefficients between the indicators were calculated by adopting the ecological engineering method applied by Li et al. in 2016. This allowed the researchers to propose the direction, principles, and mechanisms of the sustainable ecological design of regional infrastructure systems to aid in sustainable economic and social development [13]. Wang et al., 2016, established an urbanization infrastructure project sustainability evaluation index system and applied the network hierarchy analysis method (ANP) to comprehensively evaluate the sustainability of urbanization infrastructure projects [14]. In the same year, this author introduced the cloud model [14] to describe the attributes of infrastructure project sustainability, established an uncertainty measure for infrastructure project sustainability, corrected the attribute values through the penalty factor, and proposed a sustainability evaluation method for infrastructure projects based on the cloud model. Chang et al. 2018 developed an indicator system for adopting sustainability certification in a road rating system by analyzing indicators such as carbon emissions and costs [15]. Wagale et al., 2019, proposed an enrichment assessment technique based on the fuzzy preference ranking organization method, which was used to consider time and space based on the assessment of the impact of rural road construction on community quality [16]. However, these results are not fully applicable to rural road infrastructure research. Therefore, scholars such as Gopalkrishna, 2022, proposed the use of road accessibility to judge whether or not the provision of facilities in rural India is reasonable [17]. In exploring the field of rural road infrastructure evaluation, the fuzzy Delphi method has demonstrated its unique advantages as a method that integrates fuzzy mathematics and expert opinion. Previously, researchers have attempted to apply the fuzzy Delphi method when dealing with similar infrastructure assessments involving complex and uncertain factors. For example, in the assessment of water conservancy projects, transportation network planning and public facility service quality, the method effectively solves the problem of the ambiguity and uncertainty of evaluation indexes and improves the reliability and scientific robustness of assessment results through the fuzzification of expert opinions [18]. These studies not only verify the applicability of the fuzzy Delphi method, but also provide a solid theoretical foundation and practical reference for this paper to apply it to rural road infrastructure evaluation. Meanwhile, based on the in-depth analysis of the current situation in the field of rural road infrastructure evaluation, this study innovatively integrates the fuzzy Delphi method with the structural equation model, aiming to construct a more comprehensive and scientific evaluation system. By precisely citing and expanding the existing theories, this study not only deepens the understanding of the combined method of fuzzy comprehensive evaluation and quantitative analysis, but also further enriches the theoretical framework in the field of infrastructure evaluation. Specifically, we utilize the structural equation model to reveal the internal logic and causal relationship among evaluation indicators, and effectively deal with the ambiguity and uncertainty of expert opinions by means of the fuzzy Delphi method, so as to realize the accurate portrayal and dynamic adjustment of the evaluation system. This work not only provides new ideas and methods for the evaluation of rural road infrastructure, but also lays a solid foundation for the further exploration and practical application of the subsequent research at the theoretical level.

We also highlight various factors that hinder infrastructure sustainability. For example, various constraints to infrastructure sustainability are identified from a management perspective. These include lack of maintenance, inadequate planning [19] and inefficient operation and management [20], or studying the sustainability of infrastructure in townships or rural areas through the provision of financial and policy support [21], and as a result, it is also concluded that socio-economic and financial challenges are responsible for the failure to provide sustainable infrastructure. Li identified insufficient funding as one of the biggest challenges to the development of infrastructure [22]. These literature reviews on the topic of infrastructure sustainability and townships provide useful entry points for constraint identification. However, there are few relevant studies that directly examine the sustainability constraints of rural road infrastructure.

The purpose of studying the sustainability of rural road infrastructure is to explore its key role in promoting rural economic, social and environmental development and to identify the main constraints to its development. By thoroughly analyzing the linkages between these constraints and sustainability, this study aims to propose effective strategies and recommendations to achieve long-term stability, efficient operation and environmental friendliness of rural road infrastructure. These objectives will be achieved through systematic analysis and empirical research using a combination of multidisciplinary theories and methods.

The sustainability of road infrastructure occupies a central place in regional development, and its importance cannot be overstated. It is directly related to transportation efficiency, resource utilization, environmental protection and the sustainability of socio-economic activities. The sustainability of road infrastructure is particularly critical in rural areas, which are subject to multiple constraints, such as complex terrain, lack of funding, technological limitations and sparse population distribution. Therefore, optimizing rural road networks and enhancing their sustainability are of great significance in promoting rural economic development, improving the quality of life of residents and realizing urban–rural integration.

2. Methodology

Shaanxi Province, China, is the subject of this study on rural road infrastructure constraints, mainly because of the diversity of its geographic, economic and social characteristics, which are representative of the complexity of China’s vast rural areas, and provide a wealth of cases and data support for in-depth research. The dataset used in this paper comes from a survey on the topic of rural western China. The authors chose 101 sample villages distributed in the southern Shaanxi region, the Guanzhong Plain, and the northern Shaanxi region of Shaanxi Province. This set of sample villages can be regarded as representative of almost every region because of the variety of terrain and landforms included. In January 2023 and December 2023, the working group visited townships and villages twice to collect data (

Table 1. Statistics on rural road infrastructure in Shaanxi Province.

Regional Division	Prefecture Level City	Number of Villages Surveyed	Number of Roads	Sample Size by Number of Roads			Average Main Road Width (Meter)
				≤1	(1, 5)	≥5
Southern Shaanxi	Hanzhong	13	28	1	12	0	3.5~4.5
	Shangluo	9	17	2	6	1	3.6~5
	Ankang	8	15	1	7	0	4~5
Guanzhong Plain	Xi’an	5	26	0	1	4	6~18
	Xianyang	7	26	0	5	2	4~7
	Baoji	12	30	2	9	1	3~6
	Weinan	10	37	1	7	2	5~7
	Tongchuan	8	26	2	6	0	4~6
	Yangling	8	29	1	6	1	6~8
Northern Shaanxi	Yulin	9	21	2	5	2	6~8
	Yan’an	12	23	2	9	1	5~7.8

). In each study, the team surveyed the township government, village committee members and resident villagers, and assessed all rural road projects.

2.1. Defining the Indicator System

As a first step, a critical review of the published academic literature, official documents and authoritative websites was conducted to identify initial constraints to the sustainability of rural road infrastructure. As a result, a set of constraints affecting infrastructure sustainability were derived, as shown in Table 2.

In the second step, interviews were conducted to select key constraints. Semi-structured interviews are a widely adopted research methodology used to collect research data in an efficient manner [23]. They provide flexibility within questions in order to obtain in-depth information related to the actual phenomena.

Table 2. Candidate list of constraints to the sustainability of rural road infrastructure.

Serial Number	Main Contributions	Reference
1	Complex topography is not conducive to infrastructure construction	Li et al., 2022 [24]
2	The diversity of geological formations will affect the safety of the infrastructure	Zhu et al., 2022 [25]
3	Development and construction costs have a direct impact on its sustainability	Chen, 2022 [26]
4	There is a lack of funding for construction due to regional development	Bueno et al. 2015 [27]
5	Local government financial support may not be sufficient to meet sustainability needs	Yuan et al. 2019 [28]
6	The presence of fewer local sources of financing may not be able to support infrastructure development	Wang et al., 2018 [29]
7	Remote areas are generally economically weak	Wang, 2018 [30]
8	Rural economic decline leads to an inability to support sustainable infrastructure development	Wang et al., 2022 [31]
9	Rural infrastructure makes it difficult to attract investors	Wang et al., 2016 [32]
10	Rural resources for infrastructure sustainability are relatively scarce	She et al., 2022 [33]
11	Rural development potential is generally low	Gao et al., 2018 [34]
12	Inadequate management capacity of rural administrations	Shen et al., 2011 [35]
13	Rural infrastructure maintenance regulations are relatively weak	Yan et al., 2014 [36]
14	There is a lack of incentive in rural areas	Wong et al., 2017 [37]
15	Participation of residents in infrastructure sustainability is low	Wong et al., 2017 [37]
16	Communes pay little attention to rural infrastructure development	Huang et al., 2023 [38]
17	Infrastructure do not have appropriate management or are low-efficiency	Shen et al., 2011 [35]
18	There is a lack of motivation of residents for infrastructure sustainability	Wong et al., 2017 [37]
19	There is inadequate maintenance of infrastructure in rural areas	Xie, 2011 [39]
20	The wages for labor to build and operate infrastructure in rural areas are low	Wong et al., 2013 [40]
21	Rural infrastructure construction technology is relatively underdeveloped	Zhou et al., 2023 [41]
22	Early planning for infrastructure in rural areas is inadequate	Sima et al., 2011 [42]
23	There is inadequate awareness of sustainable development in rural areas	Liu et al., 2023 [43]
24	There is a low level of education in rural areas	Hannum et al., 2003 [44]
25	The number of permanent residents in rural areas is constantly changing	Zhang et al., 2018 [45]

Table 2. Candidate list of constraints to the sustainability of rural road infrastructure.

Serial Number	Main Contributions	Reference
1	Complex topography is not conducive to infrastructure construction	Li et al., 2022 [24]
2	The diversity of geological formations will affect the safety of the infrastructure	Zhu et al., 2022 [25]
3	Development and construction costs have a direct impact on its sustainability	Chen, 2022 [26]
4	There is a lack of funding for construction due to regional development	Bueno et al. 2015 [27]
5	Local government financial support may not be sufficient to meet sustainability needs	Yuan et al. 2019 [28]
6	The presence of fewer local sources of financing may not be able to support infrastructure development	Wang et al., 2018 [29]
7	Remote areas are generally economically weak	Wang, 2018 [30]
8	Rural economic decline leads to an inability to support sustainable infrastructure development	Wang et al., 2022 [31]
9	Rural infrastructure makes it difficult to attract investors	Wang et al., 2016 [32]
10	Rural resources for infrastructure sustainability are relatively scarce	She et al., 2022 [33]
11	Rural development potential is generally low	Gao et al., 2018 [34]
12	Inadequate management capacity of rural administrations	Shen et al., 2011 [35]
13	Rural infrastructure maintenance regulations are relatively weak	Yan et al., 2014 [36]
14	There is a lack of incentive in rural areas	Wong et al., 2017 [37]
15	Participation of residents in infrastructure sustainability is low	Wong et al., 2017 [37]
16	Communes pay little attention to rural infrastructure development	Huang et al., 2023 [38]
17	Infrastructure do not have appropriate management or are low-efficiency	Shen et al., 2011 [35]
18	There is a lack of motivation of residents for infrastructure sustainability	Wong et al., 2017 [37]
19	There is inadequate maintenance of infrastructure in rural areas	Xie, 2011 [39]
20	The wages for labor to build and operate infrastructure in rural areas are low	Wong et al., 2013 [40]
21	Rural infrastructure construction technology is relatively underdeveloped	Zhou et al., 2023 [41]
22	Early planning for infrastructure in rural areas is inadequate	Sima et al., 2011 [42]
23	There is inadequate awareness of sustainable development in rural areas	Liu et al., 2023 [43]
24	There is a low level of education in rural areas	Hannum et al., 2003 [44]
25	The number of permanent residents in rural areas is constantly changing	Zhang et al., 2018 [45]

In this study, a group of 15 industry practitioners were selected as respondents from a variety of backgrounds, including 5 practitioners working in county building authorities, 6 working in rural township governments, 2 from a municipal corporation and 2 senior university researchers. All of these respondents had at least 10 years of practical or research experience in the field of building management in mountain townships and were willing to participate in this study. Individual semi-structured interviews were conducted with all fifteen interviewees. A list of candidate constraints was provided to the experts, and they were invited to comment on the suitability of the listed constraints for examining rural road infrastructure sustainability.

2.2. Key Indicators

Through semi-structured interviews, constraints could be identified but were numerous and confusing, limiting the effectiveness of exploring the relationships between these constraints and identifying the main barriers to the sustainability of infrastructure in mountain townships. The use of fuzzy Delphi method (FDM) [46] was motivated by its strengths in effectively dealing with uncertainty, subjectivity and the diversity of expert opinions. By combining the reliable observations obtained from the Delphi method with the quantitative ability of fuzzy set theory to quantify uncertainty, the FDM is able to systematically collect, collate, and analyze expert perceptions and predictions of key factors in situations of limited or uncertain data, providing strong support for the development of scientifically sound strategies. The application of this methodology can help to assess the sustainability of rural road infrastructure more comprehensively and provide a reliable basis for its optimization. Therefore, in this study, the FDM was used in to screen for indicators of building service life. The process is shown below.

(1)

Create a hierarchy for each level.

(2)

Establishment of the comparison matrix. Establish a pairwise comparison matrix by comparing the relative importance of expert K’s opinion on any two evaluation indicators, i and j, in the L + 1 level in the questionnaire results: $A,A=\left[a_{ij}\right]$.

(3)

Develop a triangular fuzzy number model. The steps of the algorithm are as follows:

$${\tilde{a}}_{ij}=\left(\left[{\alpha }_{ij},{\delta }_{ij},{\gamma }_{ij}\right]\right),i,j=\mathrm{1,2},\cdots ,n$$

(1)

$${\alpha }_{ij}=\mathrm{Min}\left(B_{ijk}\right),k=1,2,\cdots ,n$$

$${\delta }_{ij}={\left({\displaystyle \prod _{k=1}^n{B}_{ijk}}\right)}^{1/n}$$

$${\gamma }_{ij}=\mathrm{Max}\left(B_{ijk}\right),k=1,2,\cdots ,n$$

In Equation (1), ${\tilde{a}}_{ij}$ is the fuzzy triangular number; ${\alpha }_{ij}$ is the minimum value of the j sub-indicator under the i indicator; ${\delta }_{ij}$ is the geometric mean of the j sub-indicator under the i indicator; and ${\gamma }_{ij}$ is the maximum value of the j sub-indicator under the i indicator. The geometric mean, ${\delta }_{ij}$, of the triangular fuzzy number for each indicator is used to indicate the consensus of the expert group on the assessed value of the criterion, which necessitates the determination of threshold k. If $k\le {\delta }_{ij}$, the indicator is selected, and vice versa.

Integration analysis of the questionnaire using MATLAB R2022a was used to set the threshold k at 6.21. It was assumed that indicators with scores below the threshold failed to meet the expert consensus and were excluded from this study. As a result, 4 indicators and 20 sub-indicators were finally obtained.

2.3. Questionnaire Survey

The questionnaire survey was conducted to collect information on the views of local residents on the infrastructure that was relatively familiar to them, so that residents could judge the degree of influence of the factors constraining the sustainable development of the actual road projects in the local context and establish a system of indicators.

The questionnaire was distributed to 552 potential respondents. These respondents were mainly government officials, village committee members, and local residents from 50 townships in Shaanxi. Respondents were selected based on the following criteria: (1) township government officials must be familiar with the rural road infrastructure positions under Suozha; (2) members of the village committees and local residents must have witnessed the entire process of road construction; (3) they must be willing to participate in the survey. Respondents also included local design firms, construction firms, and transportation-type firms. The questionnaires were filled out during the on-site research. Of the 579 questionnaires distributed, 498 valid questionnaires were completed and returned. The sample size met the statistical requirements.

Table 3. Table of questionnaire respondents.

Targets	Research Methods	Number of Respondents	Effective Number of Persons	Effective Percentage
Residents	Questionnaire	412	379	0.92
Neighborhood councils	Visits, Questionnaires	60	59	0.98
Township government	Visits, Telephone interview	45	43	0.96
County government	Telephone interview	21	19	0.90
Enterprises	Visit, Questionnaire	28	27	0.96
Others	Questionnaire	13	8	0.62

provides an overview of the questionnaire respondents.

3. Data Processing

3.1. Model Building

Structural equation modeling (SEM) [47], is an effective method for multivariate analysis, and structural equation modeling has been widely used in many fields. It is often used to refer to the application of the covariance matrix of variables in order to analyze the relationship between variables, and usually includes one structural model and two measurement models. The structural model represents a causal relationship between potential endogenous and potential exogenous variables in the model; the measurement model consists of two parts, representing the relationship between endogenous measurable variables and potential endogenous variables, and the relationship between exogenous explicit variables and potential exogenous variables.

① The structural model is shown in Equation (2):

$$\eta =B\eta +\Gamma \xi +\zeta$$

(2)

In Equation (2), $\eta$ is the vector of potential endogenous variables; $B$ is the coefficient parameter matrix of potential endogenous variables; $\Gamma$ is the coefficient parameter matrix of potential exogenous variables; $\xi$ is the vector of potential exogenous variables; and $\zeta$ is the vector of residual terms.

② The measurement model is shown in Equation (3):

$$\left\{\begin{array}{l}X={Ʌ}_x\xi +\delta \hfill \\ Y={Ʌ}_y\eta +\epsilon \hfill \end{array}\right.$$

(3)

In Equation (3), $X$ is a vector composed of exogenous indicators, ${Ʌ}_x$ represents the relationship between exogenous indicators and exogenous latent variables, and $\delta$ is the error term for exogenous indicator $X$. $Y$ is a vector composed of endogenous indicators, ${Ʌ}_y$ represents the relationship between endogenous indicators and endogenous latent variables, and $\epsilon$ is the error term for endogenous indicator $Y$.

3.2. Modeling Steps

The basic steps of model construction [48] can be divided into four steps, model setting, model identification and estimation, model evaluation and model modification, as follows:

(1)

Model setting: According to the problem under study, a hypothetical model is proposed based on relevant theories. The hypothetical model includes the relationship between the latent variables and between indicators and latent variables.

(2)

Model identification and estimation: After proposing the theoretical model, using the relevant information collected and using structural equation modeling analysis software, model parameter estimation can be achieved.

(3)

Model evaluation: Based on the results of structural equation model estimation, the parameter estimates are analyzed to see if they are within a reasonable range, and several different types of goodness-of-fit indices are analyzed at the same time to measure the degree of model fitting.

(4)

Model modification: During the estimation and evaluation of the structural equation model, the analysis software will provide relevant information on model correction, and at the same time, based on the relevant information, test the blind spots and oversights in the process of theoretical derivation, further adjust the initial model, re-estimate the parameters and evaluate the model, and derive a better hypothetical model.

3.3. Model Construction

As can be seen from Section 2.2, the indicator system includes 4 indicators, including economic capacity, management capacity, policy support and public participation, and terrain features, with 20 sub-indicators corresponding to them. In this paper, the four indicators, including economic capacity, are selected as potential variables, and the sub-indicators are selected as explicit variables to construct the structural equation model.

3.4. Model Experiment

According to SEM principles [47], correlations between variables, factor loadings, and direct effects of connecting variables should be calculated. In applying the survey data, indicator weights were calculated using SPSS 26.0 software [49] and SEM analyses are performed using the covariance matrix and maximum likelihood method. In order to test the results of the initial model generated via SEM with the data observed in the questionnaire survey, various methods such as $x^2/f$, the goodness of fit index (GFI), the adjusted goodness of fit index (AGFI), and root mean square error of approximation (RMSEA) were used. The final initial values obtained were as follows: $x^2/f$, 2.645 > 2; GFI, 0.847 < 0.9; AGFI, 0.817 < 0.9; and RMSEA, 0.068 > 0.05. These index’s indicate that the initial model does not satisfy the recommended levels. Therefore, the initial structural equation model needs to be redefined and optimized.

3.5. Modeling Amendments

We improved the initial model according to the recommendations of the goodness-of-fit statistic measures and modification indices [50]. After we added covariance error paths between observed or latent variables, the model showed good results and all the goodness-of-fit statistic measures were found to be in accordance with the recommended levels. The results of the improved goodness-of-fit statistic measurements in the improved SEM model are shown in

Table 4. Results of model-run parameters.

Fit Parameter	Numerical Value	Standard
$x^2/f$	2.3	1.5~5.0
NFI	0.91	≥0.9
NNFI	0.92	≥0.9
CFI	0.93	≥0.9
IFC	0.93	≥0.9
RFI	0.92	≥0.9
GFI	0.92	≥0.9
RMSEA	0.04	≤0.1

.

4. Discussion

In this study, the correlation coefficient is described as a “relative importance weight” and can be expressed as Equation (4):

$$D_i=W_i/\left(W_1+W_2+W_3+W_4\right)$$

(4)

where $D_i$ is the relative importance weight of indicator $i$. $W_i$ is the weight of the $i$ indicator, where $i$= 1, 2, 3, 4. The relative importance weights of the four indicators and their rankings are calculated according to Formula (4) and shown in

Table 5. Relative importance weights for level 1 indicators.

Number	Restraint	Weight	Rankings
F1	Policy Support and Public Participation	0.21	3
F2	Management Capacity	0.27	2
F3	Economic Capacity	0.39	1
F4	Terrain Features	0.13	4

. It can be observed that the order of importance of the four indicators after weighting is F3 (0.39) > F2 (0.27) > F1 (0.21) > F4 (0.13).

Therefore, the relative importance weights of the 20 sub-indicators are calculated as shown in Equation (5):

$$C_j=W_j\times D_i$$

(5)

where $C_j$ is the relative importance of constraint $j$ and $W_j$ is the weight share of constraint $j$, where $j=1,2,\cdots ,20$, the relative importance weights of which are shown in

Table 6. Ranking of the system of indicators of constraints.

Latent Variable	Measured Variables	Weighting	Weighted Weights	Ranking	Ranking Within Indicator
Economic capacity (0.39)	Higher construction costs	0.83	0.32	3	3
	Insufficient construction funds	0.79	0.31	5	5
	Difficulty in attracting capital	0.89	0.35	1	2
	Weak economic strength	0.87	0.34	2	2
	Insufficient financial allocation	0.81	0.32	4	4
	Insufficient attention from superiors	0.76	0.30	6	6
Management capacity (0.27)	Insufficient early planning	0.87	0.23	7	1
	Inadequate management of functional departments	0.82	0.22	9	3
	Lack of sustainability awareness in government agencies	0.83	0.22	10	4
	Poor management and maintenance	0.84	0.23	8	2
	Insufficient quota of building land	0.76	0.21	12	6
	Lack of relevant laws and regulations	0.82	0.22	11	5
Policy support and public participation (0.21)	Lack of policy support	0.83	0.17	14	2
	Insufficient public participation in decision-making	0.72	0.15	17	5
	Lack of policy incentives	0.81	0.17	15	3
	Population outflow	0.89	0.19	13	1
	Lack of public awareness of sustainability	0.80	0.17	16	4
Terrain features (0.13)	More complex topography and geomorphology	0.82	0.11	19	2
	Low ecological carrying capacity	0.87	0.11	18	1
	Frequent ecological disasters	0.73	0.09	20	3

.

Table 6 shows that the most influential constraint is “Economic capacity”, followed by “Management capacity”, “Policy Support and Public Participation”, and finally, “Terrain Features”. The most important constraints in terms of economic capacity are “Difficulty in attracting capital”, “Weak own economic strength”, and “High construction cost”, which indicates that the lack of financing is a major factor in the sustainability of rural road infrastructure. This indicates that the lack of infrastructure financing is the biggest obstacle to the sustainability of rural road infrastructure. In fact, according to this study’s field survey of 101 rural villages, most of the funds for road infrastructure construction come from financial grants, even in the better-off villages. Township governments are relied on to raise other small portions of funding through measures such as land grant revenues. This is because the higher construction costs are mainly related to the topographical and ecological characteristics of the sample villages.

In addition to this, “Insufficient early planning” and “Poor management and maintenance” are the most important limitations among the management capacity constraints, and in this respect, inadequate planning and lack of knowledge of the sustainable development of buildings on the part of township authorities seriously hamper the sustainability of the road infrastructure. Sustainability is seriously hampered. “Population outflow” is the most important limitation among the policy support and public participation constraints, which suggests that there are not enough people to plan, implement, and manage sustainable infrastructure. We urgently need to propose a series of targeted strategies to address these challenges in order to promote the sustainable development of rural road infrastructure.

4.1. Improving Economic and Management Mechanisms

This study conducted an in-depth exploration of the constraints on the sustainability of rural infrastructure, and concluded that economic capacity is the most important factor affecting the sustainability of rural road infrastructure, by constructing a constraints index system and calculating the relative importance weights of each factor. Among them, difficulties in attracting capital, insufficient construction funds, economic weakness and high construction costs are particularly prominent. This indicates that the lack of funds and poor financing channels have become the key bottleneck restricting the sustainable development of rural infrastructure. Therefore, the township government should actively broaden financing channels and explore diversified investment and financing modes to alleviate the financial pressure on and promote the sustainable development of rural infrastructure construction. Meanwhile, management capacity also occupies an important position among the constraints. Insufficient early planning, inadequate management of functional departments, and poor management and maintenance have seriously affected the construction quality and utilization efficiency of rural infrastructure. Therefore, township governments should strengthen the planning leadership and improve management capacity to ensure the scientific robustness and rationality of infrastructure construction [51]. In addition, they should also strengthen the post-management and maintenance work to extend the service life of infrastructure and improve its comprehensive benefits [52].

To enhance the sustainability of rural infrastructure, township governments need to start from the economic and management aspects, not only to solve the problem of a shortage of funds, but also to improve the management capacity. Through strengthening policy guidance, broadening financing channels, optimizing the planning layout, strengthening management and maintenance and other measures [53], the sustainable development of rural infrastructure construction can be promoted.

4.2. Sound Policy Support and Public Participation

The sustainable development of rural road infrastructure is limited by the problems of insufficient policy incentives, low levels of public participation in decision-making, and population loss, which highlights the key role of policy incentives and public participation in enhancing the sustainability of rural infrastructure.

To address the problem of insufficient policy incentives, township governments should increase policy support for rural infrastructure construction and formulate more favorable policies and measures to attract more social capital to participate in rural road infrastructure construction [54]. At the same time, a sound policy implementation mechanism should be established to ensure that the policy is put into practice and effective. For the problem of insufficient public participation, the township government should broaden the channels of public participation and increase the degree of public participation. By organizing hearings, soliciting opinions, and using other means, the public’s opinions should be heard, so that they can participate in the decision-making process of infrastructure construction [55]. This will not only enhance the public’s recognition and support for infrastructure construction, but also improve the scientific robustness and rationality of infrastructure construction.

In addition, the problem of population exodus is also an important factor constraining the sustainability of rural infrastructure. Township governments should focus on developing the local economy, raising the income level of farmers, and enhancing the attractiveness of rural areas. Through the development of characteristic industries, optimization of industrial structure, and upgrading of public service level, as well as other measures, they can encourage the return of members of the population and provide strong support for the sustainable development of rural infrastructure [56].

4.3. Optimizing the Construction Layout Plan

Problems such as complex topography and geomorphology, low ecological carrying capacity, and frequent ecological disasters have increased the difficulty and costs of rural infrastructure construction, posing challenges to its sustainability. Therefore, in promoting rural infrastructure construction, the impact of topographic features must be fully considered.

In response to the challenges posed by topographical features, township governments should strengthen planning and design and promote infrastructure construction according to local conditions. In selecting sites, layouts, and design plans, they should give full consideration to topographical features and ecological factors to avoid damaging the natural environment. At the same time, eco-friendly construction methods and technical means should be actively explored to reduce the impact of infrastructure on the ecological environment. In addition, township governments should also strengthen ecological protection and restoration work to enhance the carrying capacity of the rural ecological environment [57]. Through the implementation of ecological restoration projects, the promotion of eco-agriculture, and other measures, the quality of the rural ecological environment can be improved to provide a good ecological foundation for the sustainable development of infrastructure [58].

To summarize, in order to improve the sustainability of rural infrastructure, township governments need to comprehensively consider various factors such as economy, management, policy, public participation, and terrain features. By formulating scientific and reasonable policies and measures, strengthening planning, design, management, and maintenance, broadening financing channels, and enhancing public participation and awareness of ecological protection, they can promote the sustainable development of rural infrastructure construction.

5. Conclusions

This paper examines 20 constraints that impede the sustainability of rural road infrastructure, categorised into the following four indicators: economic capacity, management capacity, policy support and public participation, and terrain features. Economic capacity constraints were found to be the critical obstacle, with the constraint of difficulty in attracting capital being particularly prominent, followed by inadequate financial allocations. The top-ranked sub-indicators in the indicators management capacity, policy support and public participation, and topographical features were as follows: inadequate early planning, population loss, and low ecological carrying capacity, respectively. These findings provide useful references for local governments in other townships and for other decision-makers to take appropriate actions to improve the sustainability of road infrastructure. For example, if the lack of financial resources is known to be important, more sources of financing should be introduced for infrastructure development and more favorable policies should be provided for rural areas to help improve the sustainability of road infrastructure.

Rural road infrastructure cannot be built without efficient management tools. The top rankings were “Insufficient early planning”, “Poor management and maintenance”, and “Inadequate management of functional departments”. Reasonable construction and management procedures and clear construction and management priorities are important for the construction of rural road infrastructure. Rural infrastructure construction and management work procedures have commonalities and can be standardized. This is shown in Figure 1.

On the other hand, the unsustainable demand for road infrastructure caused by unstable populations also hinders the sustainability of infrastructure [59], suggesting that township governments should focus on the development of core competitive industries and specialty industries to improve the attractiveness of talents. “Low ecological carrying capacity” [60] is the most important limitation among the topographic feature constraints, and many rural villages are in the Qinling ecological reserve [61], so large-scale development of road infrastructure may cause damage to the pristine appearance of mountainous areas. Therefore, natural resources such as ecological landscapes and other special features should be fully utilized to enhance the integration of road infrastructure development with local natural landscapes and local cultures.

This research not only enriches the theory of sustainable infrastructure development, but also reveals the unique challenges of road construction and maintenance in rural environments by analyzing the constraints of management capacity, policy support and public participation on the sustainability of rural roads. The methodological rigor and in-depth empirical analysis in this study provide a framework for other developing countries or regions to learn from in improving rural transportation infrastructure, and it promotes the deepening and development of multidisciplinary, cross-disciplinary research in transportation engineering, development economics, and policy sciences.

Although research on the constraints of rural road infrastructure has made significant progress in revealing development bottlenecks and proposing improvement strategies, it still faces several limitations. First, the breadth and depth of data collection and analysis need to be improved, especially considering the lack of detailed data for remote and underdeveloped areas, which limits the comprehensiveness and accuracy of the study. Second, the application of research methods and models needs to be further built upon and optimized to better capture the dynamic changes and interactions between the constraints. Looking ahead, studies should focus on in-depth cross-regional and interdisciplinary cooperation, strengthen data sharing and integration, and at the same time explore the potential and application paths of new technologies and policies in solving rural road infrastructure constraints, so as to provide solid support for the development of more scientific and effective strategic planning.