Revealing the Priorities for Rural Infrastructure Maintenance Through Complex Network Analysis: Evidence from 98 Counties in China

Abstract

Driven by the Rural Revitalization Strategy, China has substantially increased its investment in rural infrastructure. Nevertheless, widespread issues such as underutilization and inadequate management persist. Recognizing rural infrastructure as a complex and interdependent system, this study applies complex network analysis to evaluate data from 98 counties, treating each county as an analytical unit and various infrastructure types as network nodes. A rural infrastructure interdependency network is constructed to examine the interdependencies among infrastructure and the overarching structural characteristics of the system. The analysis demonstrates that the rural infrastructure network exhibits pronounced modularity, with three distinct functional clusters: (1) electricity–water–broadband internet, (2) public service infrastructure, and (3) housing–environmental governance infrastructure. Furthermore, by employing a network dismantling approach that simulates facility management failures through the progressive removal of nodes, this study identifies paved roads and electricity supply stability as critical nodes within the rural infrastructure network. The failure of these infrastructures triggers systemic fragmentation and functional collapse, indicating their pivotal role in maintaining overall network integrity. These findings offer theoretical support for the optimization of infrastructure maintenance strategies, with the ultimate goal of enhancing the overall resilience and sustainable development capacity of rural infrastructure systems.

1. Introduction

Rural infrastructure comprises a wide array of essential services and facilities that are critical to the functioning and development of rural communities. Adequate infrastructure plays a fundamental role in promoting rural economic growth, enhancing quality of life, and ensuring access to basic public services [1]. Globally, improving rural infrastructure is widely regarded as a fundamental strategy to narrow the urban–rural divide and advance rural revitalization. The European Union has promoted the integration of transport, energy, and digital infrastructure through its Rural Development Policy [2]; India has pursued large-scale rural electrification and public health infrastructure [3,4]; and, in Sub-Saharan Africa, the Programme for Infrastructure Development in Africa (PIDA) has sought to improve rural road connectivity, water access, and energy availability [5]. These cases reflect the heightened attention paid to rural infrastructure in global governance practices. In recent years, the Chinese government has significantly increased its investment in rural infrastructure, with expenditures amounting to CNY 1.1503 trillion in 2020—representing a 25.47% increase from 2017 and an average annual growth rate of 3.79% [6]. These capital investments have established a robust material foundation for rural development, aiming to address inequalities and spatial disparities exacerbated by rapid economic expansion [7].

However, many rural areas continue to face challenges in the operation, management, and maintenance of infrastructure during the post-construction phase. Persistent issues such as “built but unused” and “built but malfunctioning” remain widespread [8]. The underlying causes of these issues are multifaceted. Poor policy implementation often leads to a disconnect between stated goals and actual outcomes [9]. Infrastructure funding is unevenly distributed, resulting in redundancy and waste in some areas, while remote regions remain underfunded [10]. Additionally, long-term maintenance mechanisms are lacking, with unclear responsibilities and insufficient upkeep after construction hindering sustainable operation [11]. Further studies highlight the crucial role of community and social forces in sustaining infrastructure. For instance, in Mbala, Zambia, the sustainability of water infrastructure depends largely on active community participation rather than government intervention, underscoring the importance of community-driven approaches [12]. Meanwhile, the current reliance on both government and market forces in infrastructure investment, combined with diverse capital sources and insufficient coordination, often results in resource misallocation and inefficiency [13]. Such operational failures contribute to inefficient resource utilization and governance ineffectiveness, thereby undermining the practical outcomes of rural revitalization initiatives.

Extensive research has examined rural infrastructure from multiple perspectives, including spatial configuration, investment efficiency, supply–demand alignment, and equity considerations. Existing studies highlight its multifaceted economic functions: enhancing agricultural productivity [14], raising household income levels [15], expanding off-farm employment opportunities, and contributing to poverty alleviation in rural areas [16]. Beyond its economic implications, infrastructure also serves as a critical enabler of equitable access to essential public services and inclusive development. For instance, transportation and healthcare facilities directly influence service accessibility [17], while digital infrastructure plays a pivotal role in narrowing the digital divide [18]. In particular, under China’s unique policy context, the structural interdependencies among different infrastructure elements have not been fully revealed. Moreover, the existing literature predominantly focuses on single-dimensional investment–efficiency analyses, lacking the in-depth exploration of systemic resilience and the identification of critical nodes. Compared to their urban counterparts, rural infrastructure systems tend to be more fragmented and heterogeneous, characterized by high fiscal dependency and underdeveloped mechanisms for ongoing operation and maintenance. Existing research has long explored the interdependencies among urban infrastructure systems from a systems perspective [19], recognizing that infrastructure components do not function in isolation but are highly interconnected and mutually dependent. Various modeling approaches—such as system dynamics, economic modeling, and complex network analysis—have been employed to simulate and analyze these interrelationships [20]. With the increasing investment in rural infrastructure in recent years, the need for more rational and coordinated allocation has become increasingly pressing. Gutierrez-Velez [21] calls for an integrated urban–rural approach to infrastructure planning, advocating for the systematic and equitable distribution of public resources. Accordingly, adopting a systems-based perspective in the study and management of rural infrastructure is not only timely but essential.

A growing body of research has employed complex network analysis to explore the interlinkages among the Sustainable Development Goals (SDGs), offering a methodological advance over traditional composite ranking systems that often obscure the heterogeneity among individual indicators. Notably, Nilsson [22] conducted a quantitative assessment of SDG interrelations, while Pradhan [23] highlighted both synergies and trade-offs among goals. Scherer [24] identified inherent trade-offs between social and ecological dimensions, demonstrating that progress toward social targets often coincides with increased environmental burdens. Building on this, Xutong Wu [25] analyzed the temporal dynamics of SDG interactions, shedding light on their evolving interdependencies. These studies provide valuable methodological and theoretical references for the application of network-based analyses to rural infrastructure systems. Adopting a network perspective to examine the coupling dynamics among rural infrastructure components offers a novel lens through which to understand the inherent complexity of rural systems and the mechanisms underlying infrastructure coordination and integration [26,27]. These methodological insights inform our study, enabling a more nuanced analysis of the systemic coupling relationships within the multidimensional rural infrastructure indicator system and addressing the limitations of conventional methods. While inspired by SDG-related research, our study focuses specifically on rural infrastructure systems in China. In particular, SDG 6 (Clean Water and Sanitation) informs indicators on water supply and sewage treatment; SDG 7 (Affordable and Clean Energy) guides indicators on energy infrastructure; SDG 9 (Industry, Innovation and Infrastructure) relates to overall infrastructure integration; and SDG 11 (Sustainable Cities and Communities) informs housing and public service indicators. This approach draws on SDG-inspired methodologies to analyze the interdependencies among rural infrastructure components, providing a practical framework for an understanding of local system dynamics.

As a result, this study investigates the systemic interdependencies within rural infrastructure systems. We hypothesize that rural infrastructure subsystems are inherently interdependent and that the failure of key infrastructure will produce cascading effects across the entire system. We aim to identify priority elements in the maintenance of rural infrastructure using complex network analysis, thereby addressing the limitations of traditional approaches in systematically identifying critical components and providing a basis for informed policymaking. Drawing on rural settlement data from 98 sample counties in China, we develop a comprehensive indicator framework consisting of 28 variables, encompassing housing, public service, and production–living support infrastructure. Employing complex network analysis, we model the latent structural linkages among these facilities and simulate system vulnerabilities through network dismantling analysis, which emulates infrastructure malfunction due to maintenance lapses or regulatory oversight.

In this framework, the removal of specific infrastructure nodes is treated as a proxy for maintenance failure scenarios, enabling the identification of critical nodes whose absence may compromise overall system functionality. The analysis provides a conceptual foundation and illustrative evidence to guide the prioritization of rural infrastructure investments and to enhance the resilience of rural infrastructure networks. This study makes the following key contributions: (1) it develops a comprehensive indicator system encompassing three major dimensions of rural infrastructure, addressing the existing gap in understanding the overall system structure; (2) it applies complex network methods to uncover latent coupling structures and assess systemic vulnerabilities, thereby extending the methodological toolkit for rural infrastructure research; and (3) it simulates “node removal” scenarios to identify infrastructure elements with system-wide influence, providing theoretical support for improvements in investment efficiency and infrastructure management.

The remainder of this study is organized as follows. Section 2 introduces the construction of the rural infrastructure indicator system and the development of a network model, followed by the identification of critical infrastructure nodes and simulations of network degradation. Section 3 presents the results and offers a detailed discussion. Section 4 concludes the study and outlines key policy implications.

2. Materials and Methods

We established a research framework comprising network model construction, network dismantling methods, and key node identification. The framework involves four primary steps: constructing a rural infrastructure indicator system, quantifying the correlations among indicators, developing an infrastructure synergy network, and identifying critical nodes through network dismantling analysis.

2.1. Rural Infrastructure Indicator System

Rural infrastructure is widely recognized as the physical backbone of rural development, encompassing essential components such as housing conditions, transportation systems, energy supply, water and sanitation facilities, communication networks, and basic social services [28]. Guided by the principles of data accessibility, indicator representativeness, methodological rigor, and structural coherence, this study proposes a four-dimensional classification framework for rural infrastructure: housing infrastructure, environmental governance infrastructure, municipal and utility infrastructure, and public service infrastructure. For each dimension, a set of indicators is systematically developed to holistically capture the extent to which infrastructure supports rural livability and advances sustainable rural development.

Housing infrastructure constitutes not only the primary space of domestic life for rural households but also a central target of public policy intervention. In the context of China, rural housing security is a key component of the national Rural Revitalization Strategy. The status of housing infrastructure reflects both the quality of rural living and the degree to which state policies are effectively implemented—through policy instruments such as appliance purchase subsidies and structural safety inspections of rural housing. This study evaluates rural housing infrastructure across three dimensions: (1) housing quality, measured by the proportion of structurally safe dwellings, which reflects the safety and hazard resistance of rural housing structures; (2) basic functional facilities, encompassing indicators such as the proportion of dwellings with independent kitchens, water closets, water shower facilities, and air conditioning, all of which capture the convenience and comfort of residential life; (3) green and sustainable features, such as the proportion of dwellings with energy-saving measures, which indicate the extent to which rural housing aligns with national low-carbon and ecological development goals.

Rural environmental governance forms an institutional foundation for sustainable rural development and represents a key dimension for the evaluation of rural modernization. In rural settings, particular emphasis is placed on the control of pollution sources within the living environment, with a specific focus on domestic waste and sewage management. Thus, this study constructs three representative indicators: (1) the proportion of villages implementing waste classification, which reflects both the level of public participation in environmental governance and the administrative capacity of village-level institutions; (2) the proportion of households with sewage treatment, indicating the coverage and operational status of rural sewage infrastructure; (3) the proportion of villages with managed public toilets, which allows us to assess the long-term sustainability of basic public sanitation services.

Municipal and utility infrastructure serves as the material foundation for rural social and economic activities, with its completeness and operational efficiency directly shaping the convenience of residents’ daily lives and the sustainability of regional economic vitality. In line with evolving development trends, rural infrastructure is shifting from basic coverage toward high-quality operation. For this reason, evaluation approaches must also transition from a binary “presence or absence” paradigm to an emphasis on functional effectiveness. This study constructs a multi-dimensional assessment of municipal public infrastructure encompassing the following components. (1) Stability of essential utilities, including electricity supply, water supply, and broadband internet connectivity. Unlike conventional studies that often rely on coverage rates, this study adopts “supply stability” as a core metric—defined as experiencing no more than three disruptions (e.g., power outages, water cuts, or broadband instability) within the past year. These indicators, derived from household survey data, more accurately reflect residents’ lived experiences and the perceived service quality. (2) Road infrastructure and lighting, evaluated through the proportion of villages with paved main roads, the proportion of villages with streetlights on main roads, and the proportion of villages with lighting in public spaces. These indicators collectively capture rural accessibility and the safety of public environments. (3) Adoption of clean energy, measured by the proportion of households with piped gas and the proportion of households using renewable energy, which reflects the modernization and sustainability of the rural energy structure.

Public service infrastructure constitutes a fundamental pillar in ensuring rural basic functions and enhancing residents’ well-being. Its coverage and accessibility levels not only directly affect the satisfaction of rural residents’ essential needs—such as education, healthcare, eldercare, and logistics—but also critically shape regional attractiveness, the population agglomeration potential, and spatially balanced development [29]. Existing studies emphasize that improving the spatial accessibility and service availability of rural public service facilities is a key pathway to promoting social equity and strengthening the endogenous driving forces of rural development [30]. In response, this study constructs an evaluation framework for public service infrastructure centered on accessibility, prioritizing residents’ actual experiences in reaching these facilities rather than relying solely on their physical presence or administrative coverage. Relevant data are derived from household surveys in which respondents report the time required to reach various types of service facilities. Response options include over 30 min, 15–30 min, 10–15 min, 5–10 min, and within 5 min, with corresponding scores assigned from 1 to 5. Aggregated at the county level and subsequently normalized, the final scores provide a more realistic representation of residents’ perceived service availability and practical convenience. The selected indicators span three major domains. (1) Educational services, including accessibility to kindergartens, accessibility to primary schools, and the proportion of schools in the county offering remote education. The first two indicators assess the spatial coverage and reachability of basic education services, while the latter reflects the capacity of digital infrastructure to enhance educational access, in line with ongoing trends in rural educational modernization and digitalization. (2) Healthcare services, represented by accessibility to village clinics, accessibility to township hospitals, and the proportion of telemedicine institutions in the county, collectively capturing the spatial availability of medical services at different administrative levels. (3) Other essential services, including accessibility to eldercare facilities, commercial centers, bus stations, and daily stores, as well as the proportion of villages equipped with express delivery stations. These indicators together constitute a multi-dimensional, cross-sectoral service provision network that reflects the equity and livability of rural spaces. By incorporating diverse service types and cross-domain coverage, this framework offers a comprehensive assessment of rural public service infrastructure and its role in advancing spatial justice and rural quality of life.

The specific indicators and their order are as follows (

Table 1. Rural infrastructure indicators.

Dimension	Order	Indicator Name	National Average
Housing Infrastructure	N1	Proportion of dwellings with independent kitchens	0.65
Housing Infrastructure	N2	Proportion of dwellings with water closets	0.66
Housing Infrastructure	N3	Proportion of dwellings with water shower facilities	0.58
Housing Infrastructure	N4	Proportion of dwellings with air conditioning	0.39
Housing Infrastructure	N5	Proportion of structurally safe dwellings	0.76
Housing Infrastructure	N6	Proportion of dwellings with energy-saving features	0.49
Environmental Governance Infrastructure	N7	Proportion of villages practicing waste sorting	0.39
Environmental Governance Infrastructure	N8	Proportion of households with sewage treatment	0.37
Environmental Governance Infrastructure	N9	Proportion of villages with managed public toilets	0.60
Municipal and Utility Infrastructure	N10	Electricity supply stability	0.75
Municipal and Utility Infrastructure	N11	Water supply stability	0.58
Municipal and Utility Infrastructure	N12	Broadband internet stability	0.51
Municipal and Utility Infrastructure	N13	Proportion of villages with paved roads	0.96
Municipal and Utility Infrastructure	N14	Proportion of villages with streetlights on main roads	0.82
Municipal and Utility Infrastructure	N15	Proportion of villages with lighting in public spaces	0.87
Municipal and Utility Infrastructure	N16	Proportion of households connected to gas supply	0.63
Municipal and Utility Infrastructure	N17	Proportion of households using clean energy	0.50
Public Service Infrastructure	N18	Accessibility to kindergartens	0.36
Public Service Infrastructure	N19	Accessibility to primary schools	0.58
Public Service Infrastructure	N20	Proportion of schools equipped for tele-education	0.48
Public Service Infrastructure	N21	Accessibility to village clinics	0.72
Public Service Infrastructure	N22	Accessibility to township hospitals	0.52
Public Service Infrastructure	N23	Proportion of hospitals equipped for telemedicine	0.44
Public Service Infrastructure	N24	Accessibility to elderly care facilities	0.51
Public Service Infrastructure	N25	Accessibility to commercial centers	0.59
Public Service Infrastructure	N26	Proportion of villages with express delivery stations	0.78
Public Service Infrastructure	N27	Accessibility to bus stations	0.58
Public Service Infrastructure	N28	Accessibility to daily stores	0.59

).

2.2. Rural Infrastructure Network

Constructing a network model for rural infrastructure provides an important approach to analyzing interrelationships among infrastructure components. A network consists of a set of N nodes and a set of edges E connecting certain pairs of nodes. In this study, infrastructure indicators are treated as distinct nodes, and the network is constructed by calculating the Pearson correlation coefficients between indicators. Edges are retained only when the correlations are statistically significant and greater than 0.4, indicating substantial and valid positive relationships. To justify the threshold choice, we compared thresholds of 0.2, 0.3, 0.4, 0.5, and 0.6. Threshold selection filters the strength of associations among indicators, with higher thresholds producing sparser networks. The results show that increasing the threshold reduces the network density, edges, clustering, and connectivity. At 0.4, the network maintains a balanced structure, with a clustering coefficient of 0.593, density of 0.254, and 20 connected nodes. This threshold effectively removes weak correlations, avoiding redundancy without isolating nodes. Lower thresholds cause excessive edges, while higher thresholds lead to overly sparse networks and isolated nodes. Therefore, 0.4 balances connectivity and sparsity well, making it appropriate for this study.

After constructing the network, this study calculates the degree centrality and betweenness centrality of each node and performs a modularity analysis to evaluate the relative importance of different nodes and their interconnections. This analytical approach facilitates a structural understanding of the rural functional relationship network.

Degree centrality is the most direct measure of node centrality in network analysis. It reflects the extent to which a node is connected to other nodes. A higher degree indicates greater degree centrality and suggests that the node holds a more prominent position within the network. The formula for calculating degree centrality is as follows:

$$D{C}_i={\displaystyle \frac{k_i}{N-1}}$$

(1)

where $k_i$ represents the number of edges connected to node $i$, N is the total number of nodes, and $N-1$ represents the maximum possible connections for node $i$.

Betweenness centrality measures the frequency with which a node appears on the shortest paths between any two nodes, reflecting its role as an intermediary that can control or influence the information flow within the network. The formula is expressed as

$$B{C}_i={\sum }_{s\ne i\ne t}{\displaystyle \frac{n_{st}^i}{g_{st}}}$$

(2)

where s, i, and t represent nodes; $n_{st}^i$ is the number of shortest paths between s and t that pass through node i; and $n_{st}^i$ is the total number of shortest paths between nodes s and t. Modularity analysis aims to identify densely connected subgroups (clusters or communities) within the network, where intragroup connections are dense and intergroup connections are sparse, thereby revealing functional partitions of the network.

The construction of a network offers a novel analytical lens to understand the interdependencies and relationships among rural infrastructure. However, considerable uncertainty remains as to how such networks can inform concrete policy actions or intervention strategies. To address this gap, this study introduces the network dismantling approach as an approach to explore potential solutions. Analyzing system responses to external disturbances holds significant practical relevance, as it facilitates the identification of latent tipping points and the anticipation of major shifts in system structure or function [31]. In complex network analysis, nodes are interconnected and jointly sustain overall system functionality. The network dismantling method has emerged as a key analytical tool in investigating structural properties, identifying critical nodes, and assessing network resilience [32]. Its core principle lies in the sequential removal of nodes or edges, followed by an evaluation of the resulting changes in network connectivity and functionality, thereby revealing the network’s resilience to targeted attacks or systemic failures. Unlike traditional centrality-based analyses that focus on the importance of individual nodes, network dismantling emphasizes the collective impact of node sets, making it particularly suitable for evaluating systemic vulnerabilities and designing effective intervention strategies. In real-world infrastructure systems, failures at critical nodes can trigger cascading breakdowns with severe consequences. Therefore, identifying and safeguarding such vulnerable components in advance is of substantial policy and practical importance. In recent years, researchers have proposed a range of node attack strategies, including collective influence and generalized network dismantling, and, more recently, machine learning-based graph decomposition algorithms, which have significantly enhanced the accuracy and efficiency of identifying structurally critical nodes [33].

For the rural infrastructure indicator network constructed in this study, we apply the betweenness interactive [34], a relatively traditional approach that dismantles the network by iteratively removing nodes with the highest betweenness centrality and recalculating the centralities after each removal. Wandelt [35] found that this method performs relatively well for small-scale networks, despite its high computational complexity, making it suitable for our study.

As the network becomes fragmented, nodes that belong to the largest connected component can still function, whereas nodes in smaller disconnected clusters lose functionality. Therefore, for interdependent networks, only the largest mutually connected component is of primary concern [36]. The existence of a giant component is typically a prerequisite for maintaining network functionality. We focus on tracking changes in the size of the largest connected component, defined as the subgraph containing the highest number of interconnected nodes. A larger component indicates stronger network connectivity. Furthermore, when the removal of a node leads to a significant reduction in the size of the largest component, this node is identified as a critical trigger for network collapse [37].

In the context of the rural infrastructure network, network dismantling signifies a severe disruption in system connectivity and overall functionality. The original interdependency relationships among indicator nodes are considerably weakened or even lost, leading to a decline in the overall operational efficiency of the rural infrastructure system. For instance, when the linkage between housing and other infrastructure components is disrupted, the individual existence of certain indicators becomes insufficient—their ability to collectively contribute to rural development is substantially diminished, making it difficult to sustain the system’s intended functions. In practice, when a specific facility deteriorates due to a lack of maintenance, its effective service level declines, thereby impairing its collaborative functioning with other facilities. Of particular concern are those facilities whose failure or disruption would have significant repercussions for socioeconomic security; such facilities are typically regarded as critical infrastructure. Their stable performance is essential in maintaining the resilience and functional completeness of the broader rural system.

2.3. Data Source

The primary data used in this study are derived from the Rural Construction Evaluation conducted by China’s Ministry of Housing and Urban–Rural Development. This nationwide rural survey is carried out at the county level. To enhance the data representativeness, 3–4 sample counties were selected from each province, and household and village cadre questionnaires were administered within each county. The sampling process carefully considered China’s complex topography and variations in socioeconomic development. In general, each province included counties representing higher, average, and lower levels of economic development, ensuring broad geographic and socioeconomic coverage.

The raw household questionnaire data were first cleaned to eliminate missing and abnormal responses. Afterwards, village- level and household-level data were aggregated to the county level by calculating averages and proportions, forming the basis for statistical analysis. In total, data from 98 sample counties across 26 provinces in China were obtained for this study (Figure 1).

3. Results

3.1. Regional Disparities in Rural Infrastructure

Our data show that there are significant disparities in the level of infrastructure provision across rural areas in China.

Table 2. Regional differences in different dimensions of rural infrastructure.

Dimension	Northeast Region	Eastern Region	Western Region	Central Region	National Average
Housing Infrastructure	0.41	0.66	0.55	0.66	0.59
Environmental Governance Infrastructure	0.43	0.6	0.36	0.46	0.6
Municipal and Utility Infrastructure	0.68	0.76	0.65	0.76	0.7
Public Service Infrastructure	0.48	0.6	0.52	0.61	0.56

presents the regional differences in different dimensions of rural infrastructure. To facilitate visual analysis, we have plotted the difference results for different provinces in different dimensions, as illustrated in Figure 2.

In the dimension of rural housing, facility provision exhibits the most pronounced regional disparities. Currently, the rural housing levels in the central and eastern regions are at the forefront nationally, while the northeastern and western regions lag behind. The core differences primarily lie in the adequacy of interior household facilities. For example, the availability of water showers—an important indicator of housing comfort and living convenience—shows significant geographical variation. The installation rates in the eastern and central regions are 66.48% and 66.61%, respectively, which are markedly higher than 58.32% in the western region and 20.90% in the northeastern region. The situation in the northeast is particularly notable, where less than one-fifth of rural households are equipped with water shower facilities, indicating a substantial gap. This suggests that wealthier regions benefit from stronger financial capacity and policy support for housing improvement, while the northeast suffers from population decline and weakening investment incentives due to economic restructuring.

In contrast, environmental governance infrastructure shows relatively smaller regional variation. Zhejiang Province stands out due to higher standards in sewage treatment and waste classification within its sampled counties, while differences across other provinces remain limited. The western region, characterized by complex terrain and a low population density, faces ongoing challenges in providing adequate wastewater treatment and solid waste management.

Significant disparities also persist between the eastern and western regions in municipal and utility infrastructure. The central and eastern regions generally outperform the western region in the construction of water supply, road, and energy infrastructure. Notably, the data indicate that significant progress has been made in rural road hardening, with the average hardening rate exceeding 96% in the sampled counties. However, gaps remain in gas accessibility and the utilization of renewable energy, with average coverage rates of 63% and 49.8%, respectively. This indicates that municipal infrastructure development broadly correlates with regional economic strength but also reflects that this dimension has received greater national attention, narrowing the gaps.

In terms of public service infrastructure, the central region ranks highest overall, followed by the western region, while the northeastern region remains relatively weak. Specifically, the central region holds advantages in remote education, while the eastern region demonstrates optimal accessibility to primary schools. The northeastern region performs poorly in both kindergarten coverage and primary school accessibility. For healthcare services, the eastern region excels in village clinic coverage and telemedicine facilities. Both the western and northeastern regions exhibit shortcomings in telemedicine services, while the central region—despite its relatively strong foundation—still lags behind in the proportion of telemedicine coverage, indicating a need for further investment. Regarding elderly care services, both the eastern and central regions demonstrate relatively good coverage; however, the northeastern region faces a notable deficiency in elderly care infrastructure, which is structurally misaligned with its increasingly aging population. Furthermore, the eastern and central regions also surpass the western and northeastern regions in the accessibility of transportation and daily living services. The western and northeastern regions still have considerable room for improvement in public transit services. This imbalance underscores the uneven access to education, healthcare, and cultural services, often shaped by fiscal disparities and the population distribution.

These regional disparities are not merely technical or logistical but reflect the compounded influences of macroeconomic strength, institutional arrangements, and spatial constraints. For example, the northeast region’s infrastructural stagnation cannot be divorced from its long-term population outflow and declining local fiscal capacity. Similarly, the western region’s challenges are accentuated by its mountainous terrain and scattered settlements, which increase the cost and difficulty of infrastructure deployment. In contrast, the eastern and central regions benefit from higher fiscal revenues, greater policy prioritization, and agglomeration effects due to their economic centrality.

3.2. Results of the Rural Infrastructure Network

Traditional studies often examine rural infrastructure as a set of independent indicators—such as energy, education, or transportation—while neglecting the fact that these infrastructures function synergistically in practice. For instance, without a stable electricity supply, telemedicine services cannot operate; damaged roads not only disrupt logistics but also contribute to teacher attrition in rural schools. The primary objective of constructing a functional network is to uncover these latent symbiotic relationships—identifying which infrastructures are interdependent and which nodes, if failed, may trigger cascading breakdowns. Consequently, this study established a rural infrastructure network consisting of 20 core nodes, representing 71.43% of the initial indicator set, with 89 edges connecting these nodes. The results indicate that eight indicators were excluded from the initial network due to insufficient correlation. The overall connectivity structure is illustrated in Figure 3.

The network demonstrates a highly interconnected structure, indicating that rural infrastructure indicators are linked through direct or indirect pathways. For example, the stability of the rural electricity supply (N10) is directly correlated with ten other indicators, including housing conditions, sewage treatment, and broadband internet stability. While the rural infrastructure network is densely connected, these connections are not random but exhibit distinct structural patterns.

Modular analysis reveals that the network can be divided into three major functional clusters.

3.2.1. Public Service Infrastructure Cluster

This cluster focuses on the essential public services that underpin rural residents’ daily lives, encompassing eight key indicators drawn from the established evaluation framework: the proportion of villages with streetlights on main roads (N14), the proportion of households connected to a gas supply (N16), accessibility to primary schools (N19), accessibility to village clinics (N21), accessibility to township hospitals (N22), accessibility to elderly care facilities (N24), accessibility to bus stations (N27), and accessibility to daily stores (N28). Among these, the indicator “accessibility to elderly care facilities” (N24), with a node degree of 14, functions as the core hub within this cluster, reflecting its strong interconnection with other service-related indicators. This centrality implies that elderly care services are not functioning in isolation but are deeply embedded in a broader support network of healthcare, transportation, education, and retail access.

The structural pattern of this cluster underscores the high degree of integration among rural public services. Specifically, the well-being of elderly residents relies not only on the presence of elderly care facilities but also on the spatial accessibility of township hospitals (N22), village clinics (N21), bus stations (N27), and daily stores (N28). Likewise, the accessibility of primary schools (N19) illustrates how educational services are interconnected with residential service environments, revealing households’ broader dependence on comprehensive public infrastructure. Overall, this cluster represents a collaborative and spatially coordinated system of rural public services. It serves as a foundational pillar in enhancing rural livability, promoting social equity, and supporting the realization of basic functional guarantees in rural human settlement environments.

3.2.2. Housing–Environmental Governance Infrastructure Cluster

This cluster primarily comprises indicators directly associated with the quality and functionality of residential housing infrastructure, including the proportion of dwellings with independent kitchens (N1), the proportion of dwellings with water closets (N2), the proportion of dwellings with water shower facilities (N3), the proportion of dwellings with air conditioning (N4), the proportion of households with sewage treatment (N8), the proportion of villages with managed public toilets (N9), and the proportion of villages with lighting in public spaces (N15). Collectively, these indicators reflect the fundamental dimensions of comfort, hygiene, and convenience in the rural living environment.

It is worth noting that public infrastructure indicators such as sewage treatment (N8), managed public toilets (N9), and public lighting (N15) demonstrate strong nested coupling with private housing conditions. This indicates that improvements in housing infrastructure are not solely dependent on individual household investment but are deeply embedded in and contingent upon the village-level provision of public services.

The observed structural associations—particularly the close linkages among water closets (N2), sewage treatment (N8), and public toilet management (N9)—highlight the necessity of adopting an integrated and systemic approach to rural infrastructure enhancement. Such an approach should bridge the gap between household-level amenities and community-scale service provision, ensuring coordinated development that enhances both individual well-being and collective environmental health.

3.2.3. Electricity–Water–Broadband Internet Cluster

This cluster comprises three essential infrastructure indicators that have garnered significant national investment in recent years: stability of the electricity supply (N10), stability of the water supply (N11), and stability of broadband internet connectivity (N12). These indicators form tightly coupled dyadic relationships within the infrastructure network. Although this cluster encompasses a smaller set of indicators, it constitutes a foundational subsystem that underpins the integrity of the broader service network. From a functional perspective, a stable power supply (N10) is indispensable for residential functionality, the operation of public services, and the facilitation of information exchange. Meanwhile, water supply stability (N11) and broadband internet connectivity (N12) rely on and are intrinsically linked to the power network. In effect, the electricity supply serves as the central hub through which water and digital communication infrastructure is interconnected with the wider set of service nodes.

To further explore the systemic relationships among different infrastructure subsystems, we conducted a quantitative analysis of the interaction strength among the identified clusters. We calculated the intercluster association strength by summing the standardized edge weights between clusters. The results indicate that intercluster correlations are generally weaker than intracluster correlations, suggesting relatively independent internal dynamics within each cluster. However, significant variation exists among the three clusters. The strongest association was found between the housing–environmental governance infrastructure cluster and the public service infrastructure cluster (association strength = 12.2661), reflecting their close synergy in terms of service delivery objectives and spatial coverage. In contrast, the electricity–water–broadband cluster shows weaker connections with the other two clusters, indicating that it primarily serves as a foundational layer that supports the operational environment for other infrastructure types.

In order to further identify the role of different rural infrastructures in the overall system, we utilized two core topological indicators from complex network analysis: the node degree and betweenness centrality (Figure 4). The node degree reflects the number of direct connections between a given indicator and other indicators, representing its direct impact range within the system. The analysis results show that three indicators—the proportion of households with air conditioning facilities (N4), the proportion of public spaces with lighting (N15), and the accessibility of elderly care facilities (N24)—all have a node degree of 15, ranking highest among all nodes. This indicates that these indicators are significantly positively correlated with the largest number of other indicators within the network, suggesting that they not only directly reflect the quality of life of rural residents but also play a pivotal role in driving the coordinated development of other infrastructure and services.

3.3. Network Dismantling of Rural Infrastructure

This study employed the betweenness interactive (BI) attack algorithm to conduct targeted attack experiments. The BI algorithm iteratively removes the node with the highest betweenness centrality while continuously updating the network’s topological parameters. As shown in Figure 5, the collapse process of the network was quantified by tracking changes in the size of the giant component during node removal.

The change in the size of the giant component reflects the dismantling process of the network. As nodes were progressively removed, the size of the giant component declined from its initial value of 20. Moreover, the decline followed a non-linear pattern, with different nodes exerting varying impacts upon their removal. When 12 indicators were removed, the size of the giant component dropped to 2, indicating that only fragmented, pairwise connections remained in the network. At this stage, the network was essentially broken, and the functional synergies of rural infrastructure were severely impaired. During the dismantling process, the removal of two specific indicators caused substantial declines in the size of the giant component.

N10: Electricity supply stability. Its removal reduced the size of the giant component from 20 to 17, a decrease of 13.6%, and resulted in the disconnection of N11 and N12 from the network. Electricity stability is a fundamental prerequisite for modern rural life and production, serving as a key measure of the rural infrastructure capacity. A stable electricity supply not only enhances residents’ daily life experiences but also supports industrial activities such as agricultural processing and rural e-commerce. As an indispensable part of daily life, its degradation exerts a systemic impact across the entire infrastructure network.

N13: Village road paving. The removal of this indicator caused the giant component’s size to drop from 12 to 6, fragmenting the network into two smaller components and substantially reducing the overall connectivity. Roads function not only as physical connectors but also as conduits for information, resources, and services. Road paving is particularly critical for remote and underdeveloped villages, helping to close accessibility gaps between rural and urban areas and promote social equity. The breakdown of road-based connectivity between core villages disrupts public service delivery, fragments ecological governance, and ultimately undermines the rural system’s capacity for self-repair.

Although these two key functions belong to distinct domains, both are central to rural infrastructure modernization, improvements in quality of life, and ecological sustainability. Road paving addresses physical connectivity and lays the foundation for rural development, while power supply stability signifies progress toward modernization, reflecting dual concerns for both residents’ welfare and environmental protection. Their synergy fosters a positive cycle of “infrastructure improvement–economic development–quality of life enhancement”, thereby offering comprehensive support for sustainable rural development. Both domains represent critical areas requiring focused maintenance and protection.

4. Discussion

This study adopts complex network analysis to identify critical nodes within rural infrastructure systems and to uncover the underlying interdependencies among different infrastructure types. The results reveal a high degree of structural coupling at the system level, with road and electricity infrastructure serving as central hubs. Their malfunction or removal could trigger significant systemic vulnerabilities and potentially disrupt the overall coordination of rural infrastructure networks. The discussion proceeds in two parts: first, an interpretation of the key findings is provided; second, policy implications are proposed based on the results.

First of all, the study identifies electricity and road infrastructure as pivotal components within the rural infrastructure network. The existing literature across various disciplines has consistently underscored the strategic importance of these two sectors. As a cornerstone of rural modernization, a stable and reliable electricity supply not only underpins basic quality of life but also plays a vital role in advancing industrial upgrading, promoting educational equity, improving access to public services, and enhancing environmental sustainability [38].

In economically disadvantaged rural regions, the availability of affordable and consistent electricity has been shown to significantly improve agricultural productivity, facilitate mechanization and digital transformation in farming practices, and thereby contribute to poverty alleviation [39]. Electrification has also been linked to marked improvements in healthcare delivery [40] and in educational outcomes, particularly through enhanced learning environments and expanded access to remote education [41].

Furthermore, Minas [42] highlights that the uneven spatial distribution of energy infrastructure between urban and rural areas is a major driver of regional disparities. Strengthening rural energy systems, therefore, is essential in narrowing the urban–rural divide and promoting balanced regional development. In the context of accelerating rural digitalization and the emergence of new rural industries, the significance of electricity infrastructure continues to grow [18].

Road infrastructure plays a critical connectivity function within both the rural infrastructure network and the broader socioeconomic landscape. It serves as a foundational enabler for market access, the delivery of essential public services, and the creation of employment opportunities in rural areas. Well-developed rural roads are instrumental in enhancing agricultural productivity and supporting the development of logistical supply chains, thereby strengthening the integration of rural producers into regional and national markets. Rural transportation is particularly vital in low-income settings, where improved road access significantly enhances the provision of public services [43]. As Démurger [44] demonstrates, efficient transport infrastructure contributes to overall regional economic performance and helps to mitigate urban–rural and intraregional disparities.

In the current push for integrated rural development—particularly the convergence of primary, secondary, and tertiary industries—transport infrastructure acts as a precondition for capital inflow, the return of skilled labor, and effective market linkages [45]. Moreover, road connectivity is indispensable for the operation of essential public service systems, enabling the delivery of waste management, sanitation, and other infrastructure-related services. It also forms the physical backbone supporting the accessibility of core public services such as education, healthcare, and social protection in rural settings.

Indeed, the upkeep of rural roads and electrical infrastructure remains critical worldwide. Gibson and Olivia [46] used data from 4000 Indonesian rural households to show that the quality of roads and electricity significantly influences both employment levels and income in non-farm enterprises. Furthermore, a broad body of literature underscores that advancing road and power infrastructure has played a pivotal role in reducing the economic development gap between rural and urban areas. From a policy standpoint, these findings offer the following recommendations to enhance the planning and management of rural infrastructure.

First, it is necessary to prioritize key infrastructural assets’ stable operation. To prevent systemic disruption, assets such as an uninterrupted power supply and high-quality roadways must be treated as the foremost targets of protection. Accordingly, they should receive preferential support in terms of resource allocation, fiscal investment, and technical assistance.

Second, it is necessary to enhance collaborative management across facilities. The current governance of rural infrastructure is fragmented across departments and operates in sectoral silos; however, this study simulates and confirms the high degree of structural coupling among disparate facilities. Therefore, institutional reforms are needed to establish joint oversight across departments and cross-sectoral coordination, creating an integrated, full-life-cycle management framework from construction and operation through renewal.

Third, it is necessary to strengthen whole-life-cycle maintenance mechanisms. By modeling facility failures due to post-construction maintenance gaps, this study underscores the critical impact of ongoing upkeep on overall system stability. It is recommended that robust monitoring systems and rapid response mechanisms be implemented and that “integrated build-and-manage” policies be advanced to ensure that these facilities deliver sustained benefits over the long term.

Although this study attempts to reveal the collaborative structure and critical nodes of rural infrastructure from a complex network analysis, several limitations remain.

First, data limitations may affect the precision of the analysis. While we have constructed a relatively comprehensive indicator system, the data are primarily derived from surveys and publicly available statistics. Dynamic information such as the operational status and service quality of certain facilities is difficult to capture. Moreover, the selection of indicators inevitably involves subjective judgment, which may lead to the omission of facility elements that play important structural roles.

Second, the construction of the complex network is based on a correlation matrix, which fails to fully reflect the causal relationships or resource flow paths between facilities. Future studies could incorporate causal inference methods or spatial weight matrix models to further explore the interaction mechanisms and evolutionary logic among infrastructure components.

Finally, this study focuses on a static network structure and does not consider the dynamic evolution and intervention responses of the infrastructure system over time. Future research may build multi-temporal or panel data networks to simulate the degradation–intervention–recovery processes of infrastructure systems, thereby enhancing the predictive power and decision support value of the analysis.

5. Conclusions

This study uses rural data from 98 sample counties in China to construct a 28-indicator rural infrastructure network and analyze its systemic properties. Through network decomposition, we simulate failure scenarios resulting from inadequate maintenance, thereby uncovering the network’s structural characteristics and identifying its critical nodes. We find that rural infrastructure components are highly interlinked, with the power supply and road networks occupying central positions that far exceed the influences of other nodes. Their stability is essential to the coordinated functioning of the entire system and thus warrants prioritized maintenance.

This study systematically reveals the intrinsic connections among subsystems of rural infrastructure from a complex network perspective, overcoming the limitations of traditional single-indicator or fragmented analyses. Theoretically, our research breaks through conventional component-based evaluations by introducing a comprehensive network framework that reveals the holistic nature of rural infrastructure systems and the interdependencies among nodes. This approach enriches the theoretical framework of systemic rural infrastructure research and provides novel methods for the identification of critical nodes and assessment of network resilience. Methodologically, the use of network dismantling to simulate facility failure offers an innovative approach to understanding infrastructure vulnerability and enhances our comprehension of system resilience mechanisms.

Practically, our findings provide specific guidance for rural infrastructure planning and management by emphasizing the dual importance of construction and maintenance. Unlike traditional balanced investment strategies, this study advocates for strategic resource allocation focused on core nodes—specifically the power supply and road networks—which occupy central hub positions and exert influences far beyond those of other components. Based on the network structural characteristics, we recommend (1) prioritizing the operational assurance of key nodes to prevent systemic risks triggered by weak links; (2) enhancing the coordination between infrastructure clusters to strengthen overall system coupling; (3) considering regional differences in infrastructure development and maintenance strategies; and (4) promoting a paradigm shift in rural infrastructure governance from “construction-oriented” to “balanced construction and maintenance”. These evidence-based policy suggestions aim to optimize investment priorities, improve overall system effectiveness and resilience, and provide a scientific reference for the formulation of differentiated management strategies across diverse rural contexts.

In summary, this research offers theoretical and policy insights to pinpoint critical assets, optimize investment priorities in rural development, and strengthen the long-term resilience of infrastructure systems.