How Smart City Pilots Succeed—Based on the Qualitative Comparative Analysis of Fuzzy Sets of 35 Cities in China

Abstract

In China, smart city pilots has become an important scheme to promote the modernization of the national governance system and capacity. Based on the TOE framework, this study takes 35 Chinese smart cities as sample cities and uses the qualitative comparative analysis (QCA) to explore the influencing factors of smart city pilot construction. The results show that: (1) No single factor can constitute the necessary conditions for the high and non-high construction efficiency of a smart city pilot. (2) There are five configurations leading to the high construction efficiency of a smart city pilot, which can be summarized into three driving modes: the organizational mode composed of organizations, the organization–environment mode composed of organizations and environment, and the technology–environment mode composed of technology and environment. There are three driving modes of non-high construction efficiency of a smart city pilot, which have an asymmetric relationship with the driving mode of high construction efficiency of a smart city pilot. (3) There is a potential substitution relationship among relevant factors in the aspects of technology, organization, and environment, which can effectively replace and promote the efficient construction of a smart city. The research results have implications for improving the effectiveness of smart city construction and promoting urban innovative development and sustainable development.

1. Introduction

In 2014, the National Planning for New-Type Urbanization (2014–2020) was published, which explicitly proposed the promotion of smart city pilot construction. It outlined six key directions, including “promoting broadband information networks, the informatization of urban planning and management, intelligent infrastructure, convenient public services, the modernization of industrial development, and refined social governance”. The report of the 20th National Congress of the Communist Party of China further emphasized the need to “implement urban renewal actions, strengthen urban infrastructure construction, and build smart city pilots”. Currently, there is no unified definition of a smart city in academic circles both domestically and internationally. Some scholars argue that “technology” is the key element in smart city pilot construction [1], while “knowledge” constitutes the essence of its “smartness” [2]. Some scholars integrate different perspectives, suggesting that smart city pilots are investments in human capital, social capital, and modern communication technology infrastructure [3], forming an “integrated symbiotic intelligence” through the intelligent integration of urban “cyber–physical-social systems” [4]. As “a strategy or a policy construction method” [5], smart city has become an important carrier for urban governance innovation in China [6] and a significant solution for promoting the modernization of China’s national governance system and capabilities [7].

Before the implementation of the “Interim Measures for the Administration of National Smart City Pilot Projects” in 2012, the construction of smart city pilots in China was primarily driven by local autonomous exploration. After the release of this document, the Ministry of Housing and Urban–Rural Development officially launched the national smart city pilot program, marking the beginning of a national-level exploration into smart city pilot construction. Between 2013 and 2015, three batches of national smart city pilot lists were announced, with a total of 290 cities participating. Driven by a succession of policies, the wave of smart city pilot construction in China has become unstoppable. Currently, China is the world leader in the quantity and scope of smart city development projects [8]. However, there are still significant differences in the construction levels among pilot cities [9]. The effectiveness of smart city pilot construction is influenced to varying degrees by a number of factors, including the inherent needs of urban development [9,10], policy support [9,11], political support [12,13], administrative leadership [12], intergovernmental pressure [10,14], culture [12], technological infrastructure [14,15], economic development level [10], and urban openness [16]. Existing research on the factors affecting smart city pilot construction effectiveness often focuses on the “net effect” of single factors. However, smart city pilot construction is a complex system engineering project driven by the coupling of multiple factors, and the dominant driving factors often vary across different cities [17]. How do multiple factors couple to influence the effectiveness of smart city pilot construction? What are the differences in the dominant driving factors among different smart city pilots? This study selects 35 smart city pilots, including provincial capitals, municipalities directly under the central government, and municipalities with Independent Planning Status under the National Social and Economic Development, as research objects. Using the fuzzy-set qualitative comparative analysis (fsQCA) method, the study explores the “configurational effects” behind the construction of a smart city in China. It further summarizes and refines the scalable driving paths to enhance the construction level of a smart city, providing more specific ideas for the development of smart city pilots in China.

2. Literature Review

With the advancement of communication technology, the scientific development of cities has led to the gradual improvement of smart city solutions, and academic research on smart city pilots has become more diverse and in-depth. At present, smart city research mainly focuses on the construction of evaluation systems, policy effectiveness assessment, and the study of influencing factors.

Since the commencement of smart city pilot construction in China, the establishment of a scientific evaluation index system and the evaluation of smart city development levels have become important issues of concern in academic circles. Based on the ANP-TOPSIS method, Xiang and Ren proposed that smart city evaluation should be conducted from three levels: economy, technology, and society. They also suggested using five evaluation factors: urban infrastructure, urban public management services, urban information and economic development, urban humanistic and scientific literacy, and urban residents’ subjective perceptions [18]; Wang and Duan used the analytic hierarchy process (AHP) to construct an evaluation index system from four levels: infrastructure, public management applications, public service applications, and public support systems [19]. In 2016, the “Evaluation indicators for new-type smart cities” (GB/T33356-2016) was jointly released by the National Development and Reform Commission, the Office of the Central Cyberspace Affairs Commission, and the National Standardization Administration, becoming the core basis for the evaluation of new smart cities nationwide. The State Administration for Market Regulation recently approved the release of the new “Evaluation indicators for new-type smart cities” (GB/T33356-2022), which will officially take effect on 1 May 2023. The new “Evaluation indicators for new-type smart cities” specifies the evaluation indicators system, indicator descriptions, and indicator weights for new smart cities at the prefecture level and above, including eight first-level objective indicators such as public services, precise governance, information infrastructure, and information resources, as well as one first-level subjective indicator for citizen experience, 29 s-level indicators, and 62 s-level sub-indicators. The new “Evaluation indicators for new-type smart cities” also provides reference evaluation indicators for the construction of new smart city pilots in counties and county-level cities.

As an urban governance program to solve the problem of city disease and promote sustainable urban development, “smart city” has been empirically proven to have positive impacts on various aspects, including enterprise total factor productivity [20], business environments [21], urban low-carbon development [22], environmental pollution control [23], and the reduction of urban–rural income gaps [24]. Smart city pilot construction can promote high-quality economic growth [25], enhance citizen participation, and effectively improve urban management performance and public safety levels [26]. However, some scholars have also pointed out the potential risks and shortcomings of smart city pilot construction. The process of smart city pilot construction may face various risks, including conceptual, technological, industrial, and social risks [27], as well as significant information security challenges [28]. It may also lead urban governance into the trap of “technological determinism”, exacerbating social conflicts [26].

Academic circles generally agree that the effectiveness of smart city pilot construction is influenced by multiple factors. On one hand, scholars have explored the specific mechanisms of action from different levels of smart city pilot construction. Zhang et al. [29] constructed a framework model of information security influences in smart cities, explaining the impact of four major factors—user behavior, data services, personnel management, and external environment—on smart city information security. Deng [30] assessed the development level of smart cities and explored its influencing factors through citizens’ acceptance and perception of smart public information services. On the other hand, many scholars have explored the influencing factors of smart city pilot construction from a macro perspective. Some scholars have adopted a single-factor perspective, analyzing the impact of individual factors such as technology [31], funding [32], talent [33], and citizen participation [34] on smart city pilot construction. More scholars have adopted a multi-factor perspective, analyzing the impact of multiple factors on smart city pilot construction. Leydesdorff [35] argued that the smart city is formed through the interaction of three elements: intellectual capital, industrial capital, and democratic governance, while also emphasizing the importance of information and communication technologies. Based on existing smart city evaluation index systems, Zheng and Fu [15] conducted quantitative research using five dimensions—infrastructure, technology services, economic development, industrial investment, and value realization—and 16 sub-factors, extracting the main factors influencing smart city pilot construction in western China. Yu and Xu [12] constructed a model containing five elements (policy demand, administrative leadership, political support, resources, and culture) and two rationalities (political rationality and technical rationality) to explain the development of smart cities in China, particularly emphasizing the importance of political support in smart city development. Cui et al. [36], taking Shandong Province as an example, analyzed the positive impact of governance, economy, technology, and data on the construction and operation of smart cities, although the degree of impact varied. Gazzeh used a combined content analysis and analytic hierarchy process techniques to rank the importance of sustainable smart city indicators, providing a methodological framework for understanding the relative importance of key indicators [37]. Praharaj et al. [38] developed a city typology of 100 smart cities in India based on Key Performance Indicators, which provided insights into the current status and trends, emphasizing the importance of contextualization.

In summary, the current academic research on the influencing factors of smart city construction focuses on exploring diversified influencing factors and focuses on verifying and illustrating the “net effect” of influencing factors, which to a certain extent can provide a reference for the construction of smart cities, but there are certain limitations, as the construction of smart cities is often influenced by multiple subjects and multiple factors at the same time, and each factor interacts with each other and is closely linked. Under this circumstance, it is difficult for a single “net effect” study to explain the mechanism behind the high construction effectiveness of smart cities. Existing studies mainly focus on identifying or prioritizing individual indicators, but fail to fully reveal how multiple factors combine and interact to influence the effectiveness of smart city construction; they do not sufficiently explore the specific driving paths for different types of cities to achieve the high construction efficiency of smart cities and the potential substitution relationships between factors. These studies have neglected the comprehensive impact of factors and have not proposed a driving model for smart city construction from a configurational perspective.

Therefore, this study will apply the fsQCA method to delve deeper into the configurational effects and equivalent pathways behind the effectiveness of smart city pilot construction in China, and further clarify the comprehensive mechanism by which the influencing factors of smart city construction play a role. This will help provide finer and more operational knowledge on how to drive efficient construction based on the recognition of city differences, going beyond the importance ranking of a single indicator to provide more comprehensive and in-depth insights into smart city construction.

3. Research Framework

3.1. TOE Framework

The TOE framework (Technology–Organization–Environment) was proposed by scholars Tornatzky and Fleisher in 1990 [39]. It is an essentially theoretical framework based on technology application fields rather than a purely theoretical framework. It can serve as a reference for factor selection and has broad applicability. The TOE framework includes three levels of factors: T represents technological factors, referring to the characteristics of existing and to-be-adopted technologies and the relationship between technology and the organization, that is, the degree to which technology matches the organization’s capabilities and structure, including technological resources and related capabilities [40]; O represents organizational factors, including organizational size, organizational systems (formal/informal), and the resource capabilities of the organization [41]; E represents environmental factors, encompassing various external pressures and supportive factors [42], including market environment, policy environment, and industry competition. The TOE framework was initially applied to describe the adoption of innovative technologies by organizations, and it is believed that the adoption of innovative technologies is influenced by three factors: technology, organization, and environment. Based on this, scholars have gradually expanded the framework’s applicability to other research fields. At present, the TOE framework has been widely applied in the administrative management field in China, including the enhancement of government e-service capabilities [43], digital government construction [44], government website construction [45], and poverty alleviation governance [46].

Considering the current reality and existing literature, it is evident that smart city pilot construction is influenced by diverse factors such as urban resource endowments, external environments, and the inherent technological demands of smart cities. This aligns closely with the triple logic of the TOE framework. The TOE framework simultaneously captures the combinatorial effects of technological compatibility, policy support, and superior pressure, while also revealing substitutive pathways between factors—unlike single-factor models. In China, smart city pilot programs have been characterized by a centrally directed, locally executed paradigm. Therefore, this study uses the TOE framework as a reference to establish an analysis model of the factors influencing the effectiveness of smart city pilot construction and identify key conditional variables affecting the effectiveness of smart city pilot construction. Since this study adopts the qualitative comparative analysis (QCA) method and the case is a medium sample, to reduce the impact of “limited diversity”, the number of conditional variables is controlled to 7 [47]. Specifically, the technological level includes two factors: technological infrastructure and technological human capital; the organizational level includes two factors: financial resources and political support; and the environmental level includes three factors: superior pressure, public demand, and cultural openness, forming an integrated analysis framework (see Figure 1).

3.2. Analysis Framework

3.2.1. Technological Factors

This study primarily selected technological infrastructure and technological human capital as two influencing factors at the technological level. The development of smart city pilots relies on information and communication technologies, which in turn require the hard support of infrastructure. Additionally, the implementation of smart projects such as smart transportation, smart government, and smart healthcare requires not only “hardware conditions” but also “soft conditions”, that is, the support of relevant technical talents. High levels of technological human capital play a crucial role in the development of smart cities [48]. Based on panel data from 220 prefecture-level cities, Chu et al. [49] confirmed that smart cities with high human capital endowments and strong human capital agglomeration effects tend to have higher innovation capabilities. Giffinger et al. analyzed the ranking of 74 medium-sized European cities, using cultural and educational facilities, level of qualification, and affinity to lifelong learning as core indicators [50]. They emphasized the importance of a technological infrastructure and technical human capital. Therefore, based on the current reality and relevant research literature, this study believes that the technological infrastructure and technological human capital of a city will influence the effectiveness of smart city pilot construction in the city.

3.2.2. Organizational Factors

This study primarily selects financial resources and political support as influencing factors at the organizational level.

The resource slack theory suggests that organizations with slack resources are more inclined to innovate [51]. The policy innovation diffusion theory also indicates that the diffusion of policy innovation depends on the organization’s resource endowments, with financial resources being a key internal factor in policy innovation diffusion [52]. The construction of smart city pilots requires a large number of communication infrastructure and the innovation and development of information technologies, all of which depend on the financial support of the government. The existing research also shows that government investment in smart city-related industries [15] and fiscal budgets [14] affect the level of smart city pilot construction. Therefore, this study selects financial resources as one of the conditional variables, positing that the richer the government’s financial resources, the more investment in scientific and technological innovation and infrastructure construction, and the better the effectiveness of smart city pilot construction.

Existing research shows that in the implementation of various projects, such as low-carbon city construction [53], environmental quality improvement [54], higher vocational education development [55], and government service capability enhancement [56], all is inseparable from the attention and support of local governments. When smart city pilots become a national strategy, government support, political perspectives, and the provision of public and social services emerge as major drivers of smart city development [50]. The level of government attention and support significantly influences the level of smart city pilot construction. Specifically, in the process of smart city pilot construction, “smart city pilot construction requires the government to play a key role in guiding and applying information technologies, system integration, information sharing, and problem-solving [12]”. Therefore, this study posits that the strength of political support will affect the effectiveness of smart city pilot construction.

3.2.3. Environmental Factors

This study primarily selects superior pressure, public demand, and cultural openness as three influencing factors at the environmental level.

The pressure-type system is a vivid description of the operational process of local governments in China [57]. Under the pressure-type system”, vertical contracting” leads to the hierarchical delegation of superior policies and tasks, while “horizontal competition” drives officials to compete for promotion opportunities. Under the influence of these mechanisms, government officials tend to “follow orders” [58], and the pressure from superior governments has a significant impact on the behavior of subordinate governments. As a national strategy, the Party Central Committee and the State Council have continuously proposed new requirements and deployments for smart city pilot construction, placing higher demands on lower-level governments. The extent to which local governments understand and implement superior policies significantly affects the progress of smart city pilot construction [12]. In recent years, some scholars have also studied superior pressure as one of the factors influencing smart city pilot construction [10]. Therefore, based on the current reality and existing research literature, this study includes superior pressure as one of the factors influencing the effectiveness of smart city pilot construction.

The proposal of smart city pilots is problem-oriented [10]. To address and solve various urban development problems caused by rapid urbanization, the “smart city” solution emerged. Praharaj et al. analyzed data from 100 smart cities in India and concluded that smart cities of the future should focus on digital initiatives based on objective local assessments of citizens’ behaviors and aspirations [59]. They highlighted the guiding role of public demand indicators in smart city construction. Additionally, as the Chinese government is a government that serves the people, management and service are its basic functions. The nature and administrative functions of the government determine that local governments must respond to public demands and solve related problems. The louder the public’s demand for solving urban problems, the greater the demand for smart city solutions, and the greater the pressure and motivation for local governments to construct smart city pilots. Local governments meet public demands by constructing smart transportation, smart healthcare, smart education, and smart government services [12,14]. Therefore, this study posits that the level of public demand affects the effectiveness of smart city pilot construction.

Cultural openness reflects a city’s diversity and inclusiveness [60], while citizen cosmopolitanism and open-mindedness also profoundly impact smart city development [48]. The higher the cultural openness of a city, the greater the space for cultural exchange, information exchange, and technological updates. Liu and Shi [16] empirically tested that openness affects the economic growth of smart city pilots, and the possibility of innovation increases accordingly. Therefore, this study includes urban cultural openness as one of the conditional variables and believes that urban openness directly affects the application of smart city-related technologies and the implementation of related projects, thereby influencing the overall effectiveness of smart city pilot construction.

4. Research Design

4.1. Research Method

Qualitative comparative analysis (QCA) was first proposed by sociologist Charles C. Ragin in the 1980s [61]. It is a method for solving complex causal relationships based on set theory, focusing on “configurational effects” and emphasizing “multiple concurrent causalities”, which means that different combinations of conditions may produce the same result, achieving the effect of “different” [47]. It combines the advantages of both quantitative and qualitative research methods, breaking through the limitations of traditional regression analysis methods that focus only on the “net effect” of single variables. QCA has become an important research tool in various disciplines such as sociology, political science, and management. QCA mainly includes three types: crisp-set (csQCA), multi-value (mvQCA), and fuzzy-set (fsQCA). This study uses the fuzzy-set qualitative comparative analysis (fsQCA) method to examine the influencing factors and combination paths of smart city pilot construction effectiveness.

This study applies fsQCA for three main reasons: first, there are many factors influencing smart city pilot construction, and existing research mostly focuses on the net effect of single factors. However, in reality, smart city pilot construction is influenced by multiple factors, so using QCA helps to discover the combined effects of different factors, compensating for the shortcomings of existing research. Second, this study selects 35 cases, which is a medium-sized sample, making QCA the most suitable method. Third, most of the conditional and outcome variables in this study are continuous variables, which cannot be directly simplified into dichotomous variables of 0 and 1, making fsQCA more appropriate.

4.2. Sample Selection

Considering the principles of homogeneity in the overall sample and heterogeneity among samples [47], this study selects 36 cities in mainland China, including municipalities directly under the central government, provincial capitals, and sub-provincial cities, as sample cities for smart city pilot construction. Due to incomplete data for Lhasa, it is excluded, resulting in a final sample of 35 cities (see

Table 1. Sample city list.

Serial Number	City	Serial Number	City
1	Beijing	19	Zhengzhou
2	Guangzhou	20	Xi’an
3	Shanghai	21	Fuzhou
4	Shenzhen	22	Harbin
5	Chongqing	23	Shenyang
6	Hangzhou	24	Kunming
7	Qingdao	25	Hefei
8	Chengdu	26	Nanchang
9	Nanjing	27	Changchun
10	Ningbo	28	Taiyuan
11	Changsha	29	Nanning
12	Guiyang	30	Haikou
13	Tianjin	31	Yinchuan
14	Wuhan	32	Urumqi
15	Dalian	33	Hohhot
16	Shijiazhuang	34	Lanzhou
17	Xiamen	35	Xining
18	Jinan

).

4.3. Variable Measurement

4.3.1. Outcome Variable

The outcome variable of this study is the effectiveness of smart city pilot construction. This study selects the evaluation scores of sample cities from the “2019–2020 China New Smart City Pilot Construction and Development Comprehensive Influence Assessment Results Report” released by Taihao International, Guoheng Smart City Research Team, and the Research Center for Promoting Network Development as the measurement indicator for the outcome variable. The evaluation index system of this report is highly consistent with the “New Smart City Evaluation Indicators (2016)” approved and released by the National Standards Committee in 2016. Moreover, the evaluation process adopts various methods such as questionnaires and online applications for evaluation, ensuring that the evaluation scores are scientifically sound and credible.

4.3.2. Conditional Variables

Technological Infrastructure: This study measures the level of information technology infrastructure in each sample city in 2019 by the number of internet broadband connections per 100 people. The relevant data is obtained from the “China City Statistical Yearbook 2020”.

Technological Human Capital: Drawing on the approach of Yu and Xu [12], this study measures technological human capital by the number of IT industry employees per 1000 people in the sample cities in 2019. The data is sourced from the “China City Statistical Yearbook 2020”.

Financial Resources: This study measures the financial resource abundance of local governments and their financial support for smart city pilot construction by the proportion of science and technology expenditure in the local general public budget expenditure in 2019. The data is primarily sourced from the “China City Statistical Yearbook 2020”.

Political Support: Referring to the existing literature [12], this study considers the identity of the leader of the “Smart city pilot construction Leading Group” as an indicator of the level of institutional support from local governments for smart city pilot construction. If the leader of the group is the mayor or the secretary of the municipal party committee, it is considered that the smart city pilot construction has received political support, and it is assigned a value of 1; otherwise, it is assigned a value of 0. The relevant data is obtained from city government websites and related news reports.

Superior Pressure: In this study, the sample cities are searched and filtered based on whether they appeared in the three batches of smart city pilot construction pilot lists released by the Ministry of Housing and Urban–Rural Development in 2013 and 2015, and how many times they appeared. The cities are assigned values based on the number of batches they were included in. For example, if a sample city was included in all three batches, it is assigned a value of 3. Similarly, if it was included in two batches, one batch, or not included at all, it is assigned values of 2, 1, or 0, respectively. The data is sourced from relevant documents published on the official website of the Ministry of Housing and Urban–Rural Development.

Public Demand: This study uses urban population density as a proxy variable to reflect the level of public demand for smart city pilot construction. Generally, the higher the population density, the more severe the traffic congestion and environmental pollution problems, and the greater the public demand for smart city pilot construction, hoping that the government will take measures to address these issues [12]. The data for this indicator is mainly sourced from the “China City Statistical Yearbook 2020”.

Cultural Openness: This study measures the level of cultural openness of a city by the per capita utilization of foreign investment in 2019. The relevant data is sourced from the “China City Statistical Yearbook 2020”.

The assignment descriptions and data sources for the outcome variable and conditional variables are shown in

Table 2. Assignment descriptions and data sources for outcome and conditional variables.

Types of Variables			Assignment Descriptions	Data Sources
conditional variable	Technical Factors	Technological Infrastructure	the number of internet broadband connections per 100 people	“China City Statistical Yearbook 2020”
		Technological Human Capital	the number of IT industry employees per 1000 people	“China City Statistical Yearbook 2020”
	Organizational Factors	Financial Resources	the proportion of science and technology expenditure in the local general public budget expenditure	“China City Statistical Yearbook 2020”
		Political Support	If the leader of the group is the mayor or the secretary of a municipal party committee, it is considered that the smart city pilot construction has received political support, and it is assigned a value of 1; otherwise, it is assigned a value of 0.	city government websites and related news reports
	Environmental Factors	Superior Pressure	the batch number of the city’s region included in the pilot list published by the Ministry of Housing and Urban–Rural Development, with corresponding values of 0, 1, 2, or 3	the official website of the Ministry of Housing and Urban–Rural Development
		Public Demand	urban population density	“China City Statistical Yearbook 2020”
		Cultural Openness	the per capita utilization of foreign investment	“China City Statistical Yearbook 2020”
outcome variable	the effectiveness of smart city pilot construction		comprehensive influence assessment score for the smart city pilot construction and development	“2019–2020 China New Smart city pilot construction and Development Comprehensive Influence Assessment Results Report”

.

4.4. Variable Calibration

Except for “political support”, which is a dichotomous variable, all other variables are continuous. Therefore, this study adopts the direct calibration method to calibrate the original variable data, with the calibrated set membership values ranging between 0 and 1 [62]. Specifically, this study sets three membership values: 0.95, 0.5, and 0.05, representing complete membership points, intersection points, and complete non-membership points, respectively. The anchor points for each variable’s original data are calculated using functions in Excel (as shown in

Table 3. Calibration anchor points for outcome and conditional variables.

Types of Variables		Complete Membership Points	Intersection Points	Complete Non-Membership Points
conditional variables	Technological Infrastructure	94.73	52.09	31.50
	Technological Human Capital	35.80	9.60	4.21
	Financial Resources	0.0940	0.0315	0.0075
	Superior Pressure	2.95	1.05	0
	Public Demand	0.24208	0.10430	0.03535
	Cultural Openness	1328.13	583.30	25.15
outcome variable	the effectiveness of smart city pilot construction	85.87	79.97	65.99

Source: Calculated based on membership degrees.

), and finally the calibrated data is calculated using the fsQCA 3.0 software. It should be noted that after obtaining the calibrated data, this study further adjusts the membership scores of 0.5, referring to more decimal points in the basic data and comparing the original data with the data anchor points. If the original data is greater than the anchor point, 0.001 is added; if it is less, 0.001 is subtracted. This adjustment ensures that some cases are not excluded from the analysis due to difficulty in classification [63].

5. Empirical Results Analysis

5.1. Necessary Condition Analysis

Before analyzing the conditional configurations, it is necessary to conduct a necessary condition analysis. Identifying the necessary conditions for the outcome variable helps in making appropriate logical remainder assumptions and avoids treating frequently occurring conditions as necessary conditions during the conditional configuration analysis [63]. The main metrics in the necessary condition analysis are consistency and coverage. Typically, when the consistency value is greater than or equal to 0.9, the condition variable can be considered a necessary condition for the outcome variable. When the consistency is greater than 0.8, the variable is a sufficient condition [63], meaning that the occurrence of this condition variable can lead to the occurrence of the outcome variable.

Table 4. Necessary condition analysis of conditional variables.

Conditional Variables	Outcome Variable
	High Construction Efficiency		Non-High Construction Efficiency
	Consistency	Coverage	Consistency	Coverage
Technological Infrastructure	0.660	0.710	0.634	0.652
Technological Infrastructure	0.677	0.659	0.718	0.668
Technological Human Capital	0.743	0.814	0.559	0.586
Technological Human Capital	0.622	0.596	0.822	0.754
Financial Resources	0.797	0.855	0.465	0.477
Financial Resources	0.513	0.501	0.859	0.802
Political Support	0.721	0.561	0.590	0.439
Political Support	0.279	0.416	0.410	0.584
Superior Pressure	0.834	0.731	0.659	0.552
Superior Pressure	0.489	0.600	0.679	0.796
Cultural Openness	0.727	0.846	0.430	0.478
Cultural Openness	0.551	0.503	0.861	0.751
Public Demand	0.679	0.697	0.620	0.609
Public Demand	0.619	0.630	0.691	0.673

presents the results of the single-factor necessity test for high and non-high construction efficiency. From Table 4, it can be seen that the consistency indicators for all condition variables are less than 0.9, indicating that none of the condition variables constitute necessary conditions for high or non-high construction efficiency. Furthermore, the table shows that superior pressure constitutes a sufficient condition for high construction efficiency, covering nearly 73.1% of the sample cities. Other condition variables have some explanatory power for the effectiveness of smart city pilot construction, but none alone constitute necessary or sufficient conditions. Technological human capital, financial resources, and cultural openness are sufficient conditions for the non-high construction efficiency of a smart city, covering 75.4%, 80.2%, and 75.1% of the cases, respectively. This suggests that these three condition variables have good explanatory power for the effectiveness of smart city pilot construction, and their absence significantly affects the effectiveness of smart city pilot construction. The above analysis indicates that the effectiveness of smart city pilot construction is often the result of multiple factors at the technical, organizational, and environmental levels. Therefore, the next step is to conduct a conditional configuration analysis.

5.2. Conditional Configuration Analysis

In the process of analyzing the conditional configurations for high and non-high construction efficiency, it is first necessary to set the case frequency threshold and the raw consistency threshold. The case frequency threshold is closely related to the number of cases. When the research sample size is small, the case frequency threshold is set to 1. For the original consistency threshold, QCA method experts suggest that the minimum critical value is 0.75 or 0.8 [64,65]. Based on the fact that this study has a medium sample size and a small sample size, the case frequency is set to 1; Due to the cutoff value of the “natural gap” in the consistency threshold of this study being 0.8, and referencing existing literature, this paper sets the consistency threshold to 0.8. Additionally, the PRI consistency cutoff value is set to 0.7 to avoid the simultaneous subset relationship in the truth table rows [63].

Based on this, the truth table is standardized, and three combination solutions are obtained: a complex solution, a simple solution, and an intermediate solution. The intermediate solution is usually the preferred choice for scholars in QCA research [64] because it incorporates logical remainders that conform to theory, with moderate complexity and reasonable justification. Since this study does not have necessary conditions in the necessary condition analysis, the logical remainders are all set to “present or absent”, and no assumptions are made about the logical remainders. Finally, the complex solution obtained is the same as the intermediate solution. This study adopts the QCA analysis result presentation form proposed by Ragin [62] to report the intermediate solution while referring to the simple solution to distinguish core conditions from peripheral conditions. The conditional configuration analysis of high/non-high construction efficiency of a smart city pilot is shown in

Table 5. Conditional configuration analysis of smart city pilot construction effectiveness.

Conditional Configuration	High Construction Efficiency						Non-High Construction Efficiency
	H1		H2	H3			H1			H2		H3
	H1a	H1b	H2	H3a	H3b	H3c	H1a	H1b	H1c	H2a	H2b	H3
Technological Infrastructure			$⮾$	•	$⮾$	$⮾$		•			$⮾$	$⮾$
Technological Human Capital		•	$⮾$	●	●	●	$⮾$	•	$⮾$	$⮾$	$⮾$	$⮾$
Financial Resources	●	●	●	•	•	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$
Political Support	●	●			$⮾$	$⮾$		•	$⮾$	$⮾$	$⮾$	•
Superior Pressure	•	$⮾$	●	●	●	●	$⮾$	$⮾$	$⮾$	$⮾$		•
Cultural Openness	•	•	•	•	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$	$⮾$
Public Demand	$⮾$	•	•	•	$⮾$	•	$⮾$				$⮾$	●
consistency	0.96	0.95	0.95	0.98	0.99	0.99	0.99	0.99	1	1	0.91	0.95
original coverage	0.29	0.25	0.30	0.36	0.1	0.09	0.41	0.21	0.2	0.2	0.18	0.19
unique coverage	0.08	0.02	0.02	0.02	0.03	0.02	0.09	0.02	0.06	0.06	0.07	0.04
overall consistency	0.95						0.96
overall coverage	0.61						0.64

Notes: a. “●” indicates the presence of a core condition; “•” indicates the presence of a peripheral condition; “$⮾$ ” indicates the absence of a core condition; “$⮾$ ” indicates the absence of a peripheral condition; blank indicates that the condition is irrelevant; variables appearing in both the simple and intermediate solutions are core conditions, while variables appearing only in the intermediate solution are peripheral conditions. b. In the “standard analysis” operation for high construction effectiveness analysis, since China’s smart city pilot construction is a government-driven pilot, the promotion and support from superior governments and the resources controlled by the government have a significant impact on the effectiveness of smart city pilot construction. Therefore, “technological human capital × superior pressure” and “financial resources × superior pressure” are selected as prime implicants. c. Similarly, in the “standard analysis” operation for non-high construction effectiveness, the prime implicants are selected as “technological human capital × financial resources × political support” and “technological human capital × financial resources × public demand”.

. From Table 5, it can be seen that the consistency of each combination and the overall consistency are greater than 0.9, indicating that each combination can sufficiently influence the effectiveness of smart city pilots. Moreover, the pathways influencing high and non-high construction efficiency are diverse.

5.2.1. High Construction Efficiency Configuration Analysis

Specifically, the conditional variables for high construction efficiency are analyzed. From Table 5, it can be seen that there are six conditional configurations for high construction efficiency, with an overall consistency of 0.95 and an overall coverage of 0.61, meaning that the conditional configurations can explain 61% of the cases. This proves that the condition variables selected in the previous text are effective and can sufficiently explain the differences in the effectiveness of smart city pilot construction. The configurations with the same core conditions are classified into one category, and three groups of configuration results (H1, H2, and H3) are obtained, that is, three driving modes for the high construction efficiency of smart city pilots, as shown in Table 5.

The first mode is the organizational mode, corresponding to configurations H1a and H1b in Table 5. Configuration H1a indicates that under the core conditions of high financial resources and high political support, supplemented by significant superior pressure and high urban cultural openness, even without considering technical factors and low public demand, smart city pilot construction can achieve high effectiveness. This configuration can explain 29% of the cases, with 8% of the cases being explained only by this configuration. Configuration H1b shows that even if the smart city pilot construction faces little pressure from the superiors and the superiors do not have high requirements for it, with the support of more financial resources from the pilot cities themselves and the political support of the government, it can cultivate scientific and technological talents, improve the openness of urban culture, and meet higher public needs, which can also promote the high effectiveness of smart city pilot construction. This configuration can explain 25% of the cases, with 2% of the cases being explained only by this configuration. H1a and H1b share the same core conditions, and this driving mode is named the “organizational mode”. In this mode, the effectiveness of smart city pilot construction is primarily influenced by organizational factors. The pilot city government attaches great importance to the construction of smart city pilots, and the supply of financial resources is sufficient. Under these two core conditions, even if some elements are missing or insufficient, the synergistic development of other elements can still achieve the high construction efficiency of a smart city pilot. It is worth noting that “public demand + technological human capital” has a potential substitution relationship with superior pressure. Under the same core and peripheral conditions, a high public demand and high-tech human capital level can be equivalently replaced with high superior pressure to promote the high-efficiency construction of smart city pilots in the form of ‘different paths to the same end’, which further highlights the importance of organizational factors in smart city pilot construction. The analysis results show that the “organizational mode” can explain more than half of the cases, with typical sample cities mainly located in the eastern region, such as Hangzhou, Ningbo, Qingdao, Nanjing, Shanghai, and Shenzhen. These cities have abundant local government funds, and rich financial resources, closely follow the central government’s pace, and regard smart city pilot construction effectiveness as an important part of their political achievements. They have a relatively open urban culture, the overall development elements are relatively complete, and the organizational factors play a more significant role.

The second mode is the financial resource-driven organizational–environmental mode, corresponding to configuration H2 in Table 5. Configuration H2 indicates that high financial resources and high superior pressure are the core conditions, with high urban cultural openness, high public demand, non-high technological infrastructure level, and non-high technological human capital as peripheral conditions. This combination can lead to the high construction efficiency of a smart city pilot. This suggests that pilot cities with low levels of technical infrastructure construction and a lack of technological talents, under high superior pressure, in order to meet the high demand of the public, if they can fully utilize the abundant financial resources of the organization, improve urban cultural openness, and enhance urban innovation capabilities, can still drive the high construction efficiency of a smart city pilot. This configuration can explain 30% of the cases, with 2% of the cases being explained only by this configuration. This driving mode is named the “organizational–environmental mode”. This mode means that even if a city lacks the support of high technological factors, as long as the pilot city has financial resource support and is driven by the pressure-led external environment, it can achieve the high construction efficiency of a smart city pilot through the synergistic interaction of organizational and environmental factors. Typical sample cities for this mode include Nanchang and Changsha. These cities often have high economic development levels but lack technological elements. They can compensate for these deficiencies through the synergistic development of other elements to achieve the high construction efficiency of a smart city pilot.

The third mode is the technological talent-driven technological–environmental mode, corresponding to configurations H3a, H3b, and H3c in Table 5. Configuration H3a indicates that if a pilot city faces significant superior pressure and has a high level of technological human capital, with the assistance of high financial resources, high urban openness, and high technological infrastructure levels, it can achieve the high construction efficiency of a smart city pilot. Sample cities that meet all the elements of urban development at the technical, organizational, and environmental levels can largely achieve the high construction efficiency of a smart city pilot. This configuration can explain 36% of the cases. H3b emphasizes the importance of technological human capital and superior pressure. Even if other peripheral conditions are missing, with the assistance of high financial resources, the smart city pilot can achieve high effectiveness. This configuration can explain 10% of the cases. H3c differs from H3b only in the peripheral conditions, with H3c having high public demand as a peripheral condition. Configuration H3c can explain 9% of the cases. Similarly, this driving mode is named the “technological–environmental mode”, meaning that under high external environmental pressure, if the pilot city has a high technological level, even without strong support from the local government at the organizational level, the high construction efficiency of a smart city pilot can still be achieved. Typical cities for this mode include Wuhan, Dalian, and Shijiazhuang.

5.2.2. Non-High Construction Efficiency Configuration Analysis

At the same time, it can be seen from Table 5 that there are six conditional configurations leading to the non-high construction efficiency of smart city pilots. Similarly, taking the core conditions as the classification standard, the six conditional configurations are divided into three categories to obtain the driving mode for the non-high construction efficiency of smart city pilots. The H1 mode indicates that low financial resources, low superior pressure, and low urban cultural openness in pilot cities will lead to the non-high construction efficiency of smart city pilot. The original coverage and unique coverage of configuration H1a are 0.41 and 0.09, respectively, which are higher than any other configuration. This shows that this configuration is the main reason for explaining the non-high effectiveness of smart city pilot construction. Configurations H1a and H1c show that the absence of core and peripheral conditions cannot drive high effectiveness in pilot construction, which is obvious. Configuration H1b indicates that even with the presence of technological infrastructure, technological human capital, and political support as peripheral conditions, if the pilot city has low financial resources, low superior pressure, and low urban cultural openness, high effectiveness in pilot construction will not occur, emphasizing the importance of environmental conditions. The H2 mode indicates that if a pilot city lacks the core conditions such as technical human capital, financial support, and the political support of the organization, and lacks the peripheral conditions such as external environmental pressure, it cannot drive the high construction efficiency of a smart city pilot. The H3 mode indicates that in the context of high public demand, if a pilot city lacks technological human capital and financial resources, even with high political support and high superior pressure as peripheral conditions, it cannot lead to the high construction efficiency of a smart city pilot. The H3 mode further highlights the importance of technical conditions.

In the analysis of non-high construction efficiency configurations, it can be seen that financial resources play a core role. All configurations leading to non-high construction efficiency include low financial resources, indicating the important role of financial resources in smart city pilot construction. This also provides an important reference for local governments to improve the effectiveness of smart city pilot construction. By comparing the six configurations leading to high construction efficiency and the six configurations leading to non-high construction efficiency, it can be found that the driving modes for high and non-high construction efficiency of smart city pilots have an asymmetric relationship.

5.3. Robustness Test

To verify the stability of the research results, this study conducts a robustness test by adjusting the consistency threshold. The consistency threshold is adjusted from 0.8 to 0.85, while the frequency threshold remains unchanged. The conditional configuration analysis is conducted separately for high and non-high construction effectiveness. The results of the conditional configuration analysis show that the number of configuration solutions for high construction effectiveness decreases from 6 to 5, but the internal explanatory mechanisms of the configurations remain consistent with those before the adjustment, with no substantial changes. Only the consistency and coverage of the overall solutions change slightly. The coverage of the overall solution has decreased from 0.609 to 0.572, and the consistency has decreased from 0.952 to 0.949, with no significant change. The configuration solutions for non-high construction effectiveness do not change, and the coverage and consistency of the overall solution have not changed significantly. In general, the results of this study have stronger robustness.

5.4. Case Comparison Analysis

To deeply analyze the logic of different modes (organizational, organization–environment, and technology–environment) on the effectiveness of smart city construction, this study selects Hangzhou, Nanchang, and Wuhan for case comparison. These three cities represent three typical modes, with significant differences in their development history, resource endowment, and policy environment, which provide empirical samples for understanding why certain configurations lead to better outcomes in specific contexts. The following section reveals the root causes driving the success of the three modes from the perspective of city practices.

5.4.1. Organizational Mode

A typical case exemplifying this configuration is Hangzhou. The Hangzhou government plays a “trinity” role.

As a strategic planner, it formulates forward-looking blueprints. In 2014, it launched the “Number One Project” for the information economy, and in 2018, it became the first city to propose the goal of building a “China Digital Economy First City” and reinforced its execution. As a resource integrator, it guides private capital participation through special funds while opening data resources and optimizing the business environment. For example, the “Run Once” reform stimulated market innovation. As an ecosystem builder, it introduced the “27 New Talent Policies” and leveraged local industrial strengths to attract talent, creating a virtuous cycle of “policy–industry–talent.” Additionally, by capitalizing on the robust internet industry ecosystem established by local tech giants, the government integrated cross-departmental data through projects like “City Brain” to achieve multi-domain collaboration.

The success of Hangzhou’s smart city construction fundamentally illustrates the intrinsic logic of the organization-driven mode: the government activates local market ecosystems through targeted policies, effectively transforming organizational advantages into urban governance efficacy.

5.4.2. Organization–Environment Mode

A typical case exemplifying this configuration is Nanchang. Nanchang’s practice demonstrates the synergistic effect of policy guidance and organizational execution, with its environmental advantages primarily reflected in the dual drivers of top-down pressure and public demand.

As a pilot city for the transformation and application of achievements under China’s New Generation Broadband Wireless Mobile Communication Network (03 Special Project), Nanchang was included in the Ministry of Housing and Urban–Rural Development’s first list of 90 smart city pilot projects in China. This national-level strategic opportunity enabled the city to secure policy benefits and priority access to technological resources. Under the impetus of top-down pressure, Nanchang established a high-level organizational structure, forming a leadership team for smart city construction and developing specialized implementation plans to ensure the rapid allocation of relevant resources. By introducing IoT-specific policies, the city nurtured local enterprises and initially formed an innovative industrial cluster centered on the 03 Special Project. Additionally, Nanchang strengthened regional collaborative learning by actively engaging with the Pearl River Delta and other advanced regions, adopting their cutting-edge technologies and management expertise to compensate for its own shortcomings. For instance, drawing on the experience of Hangzhou’s “City Brain,” Nanchang developed its “Smart Nanchang” integrated platform to achieve one-stop online government services.

At the level of public demand, Nanchang, as a densely populated city, faces an urgent need for smart governance in its urban development process. Driven by the strategic guidance of the 03 Special Project pilot program, Nanchang Mobile has cumulatively built over 11,000 5G base stations, ensuring seamless outdoor 5G network coverage from urban to rural areas and achieving a 100% gigabit network access capability for users in towns and above. This has formed a coordinated development framework of “city–township–rural” three-tier networks. The Nanchang mode exemplifies how late-developing cities can achieve leapfrog breakthroughs in technological infrastructure through the synergistic mechanism of “policy support–resource aggregation–public demand”.

5.4.3. Technology–Environment Mode

A typical case exemplifying this configuration is Wuhan. The distinctive feature of Wuhan’s smart city development lies in the precise two-way matching between techno-logical supply and governance demand.

Wuhan boasts leading global technological capabilities in communications, remote sensing, and artificial intelligence, supported by prestigious institutions such as Wuhan University and Huazhong University of Science and Technology, as well as leading optoelectronic enterprises. The government has established a dual-channel mechanism of “technology transfer–scenario application” to translate research strengths into governance capacity. On one hand, Wuhan focuses on building robust innovation platforms. Major scientific and technological innovation bases, such as the Wuhan National Laboratory for Optoelectronics and the China Memory Base, provide foundational support for cut-ting-edge technology R&D. Additionally, the establishment of industrial technology innovation alliances accelerates the industrialization of scientific achievements. On the other hand, Wuhan actively opens complex application scenarios to drive technological innovation through demand-oriented approaches. As a megacity and transportation hub, Wuhan faces significant governance challenges in areas such as smart transportation and public safety. The government proactively opens these scenarios, creating testing opportunities for local technologies. For example, the BeiDou-based bridge monitoring system achieves millimeter-level early warnings, significantly enhancing the safety management of urban infrastructure.

With its solid industrial foundation and strong capacity for scientific innovation, Wuhan has attracted numerous foreign investments and international collaborations. In recent years, the city’s per capita actual utilization of foreign capital has shown steady growth, reflecting not only its appeal to global investors but also its sustained potential in attracting foreign investment. Wuhan’s “technology–environment” dual-cycle mode leverages its inherent technological advantages while using urban governance demands to refine and advance technological iterations. This approach provides a valuable reference for cities with abundant technological resources but pressing governance needs.

5.4.4. Comparison of Three Modes

The practices of the three cities demonstrate that the differences in smart city development outcomes are essentially manifestations of distinct resource allocation methods and governance logics under different configuration modes.

From the perspective of vertical driving forces, in the cases of Hangzhou, Nan-chang, and Wuhan, government-led vertical initiatives form the core driving force. Whether through political support, fiscal resource allocation, or other means, the government remains the key actor in overcoming governance bottlenecks and activating collaborative mechanisms. This indicates that in the complex field of smart city governance, institutional supply and guarantees play an irreplaceable role in top-level design.

From the perspective of horizontal synergistic factors, the organizational mode of Hangzhou leverages strategic planning and resource integration to precisely activate its local market ecosystem, effectively transforming organizational advantages into urban governance efficiency. This mode suits cities with a strong economic foundation, well-developed technological resources, and mature market mechanisms. The organization–environment mode of Nanchang takes national-level strategies as guidance, combines them with urgent smart governance demands, and enhances technological development through organizational safeguards, forming a synergistic mechanism of “policy support–resource aggregation–public demand.” This mode serves as a reference for late-developing cities, particularly those with limited technological endowments but favorable resource and environmental conditions. The technology–environment mode of Wuhan relies on its robust technological reserves and open cultural environment to construct a two-way feedback mechanism between technology supply and governance demand. This mode applies to cities with abundant technological resources and pressing governance needs.

Through in-depth case comparisons, this study deepens the understanding of the complexity of smart city development: no single mode fits all scenarios. Cities should, based on their own resource endowments, development goals, and challenges, analyze the intrinsic connections and interaction mechanisms among influencing factors, and adopt a context-specific approach to selecting the optimal smart city development mode.

6. Conclusions and Discussion

Based on the TOE framework, this study selects seven influencing factors from the technical, organizational, and environmental levels to construct an analytical framework for understanding the differences in the effectiveness of smart city pilot construction. Using the fuzzy-set qualitative comparative analysis (fsQCA) method to study 35 smart city pilot cases, we drew the following conclusions:

(1)

The effectiveness of China’s smart city pilots is influenced by multiple factors. None of the seven elements, including the level of technological infrastructure, can alone constitute a necessary condition for the high construction efficiency of a smart city pilot. However, from the necessary condition analysis of high and non-high construction effectiveness, it is evident that factors such as financial resources and superior pressure play a crucial role. The smart city pilot construction should be approached systematically and holistically, avoiding the fragmentation of individual elements. It is essential to recognize the interconnectedness of these elements and to guard against falling into the “isolation” trap.

(2)

The conditional configuration analysis generates six configuration results, which can be summarized into three driving modes based on core conditions: organizational mode, organizational–environmental mode, and technological–environmental mode. These six configurations and three modes reflect the multiple pathways to achieving the high construction efficiency of a smart city pilot. Additionally, based on the configuration analysis results, there are also six conditional configurations of non-high construction effectiveness, indicating a clear asymmetric relationship between the driving modes for high and non-high construction effectiveness of a smart city.

(3)

Under specific conditions, certain technical, organizational, and environmental factors have potential substitution relationships, meaning they can equivalently drive the high construction efficiency of a smart city pilot. This provides a reference for cities lacking certain development elements to promote smart city pilot construction. For example, when local governments lack financial resources and motivation, they can adopt the technological–environmental driving mode by increasing superior government pressure and attention to drive the high construction efficiency of a smart city pilot.

Based on the above conclusions, this study has the following two recommendations:

First, the high construction efficiency of a smart city pilot is the result of multiple influencing factors. In promoting smart city pilot construction, China should not focus solely on a single element but should comprehensively consider technical, organizational, and environmental factors, adopting a multi-pronged approach. In particular, it is essential to strengthen the attention of city governments to smart city pilots, and enhance political and financial support, as organizational factors play a significant role in both high and non-high effectiveness construction. Local governments should enhance their strategic prioritization of smart city initiatives, deepen their understanding of national strategies, strengthen policy implementation, and establish sustainable fiscal mechanisms to ensure long-term success.

Second, local governments should base their smart city pilot construction on local resource endowments, adapting to local conditions and choosing the smart city construction path that is suitable for it. For example, for cities with strong political support, it is recommended to give priority to the organizational mode. Such cities should actively play the government’s ability to organize, coordinate, and guide policies, which can be led by the government to set up a special leading group for smart city construction, coordinate planning, integrate resources, and ensure that the tasks of smart city construction are advanced in an orderly manner, increase the investment in smart city technology research and development and innovation, and encourage the participation of local enterprises in the construction of the project, with an emphasis on cultivating the local smart city industry ecosystem; the government should be guided by the public’s demand. In addition, the government should be oriented to public demand and widely solicit residents’ opinions and suggestions through various channels, so that the public can become participants and beneficiaries of smart city construction.

For cities with strong organizational support but low levels of technological construction and development, it is recommended to consider an organizational–environmental mode. Such cities should actively play their advantages in organizational coordination and financial security to create a good policy environment for technology development. At the policy level, the government can set up a special science and technology innovation development fund, provide financial subsidies and tax exemptions to scientific research institutions and enterprises involved in the construction of technological infrastructure, reduce their construction and operation costs, and stimulate the enthusiasm of market players to participate in the process; formulate stringent technological standards and norms to ensure that the newly upgraded technologies are in line with the development trend of the industry and the long-term planning of the city, so as to avoid the waste of resources and duplication of construction; strengthen the cooperation with universities, scientific research institutions and related enterprises; and establish a mechanism for collaborative innovation between industry, academia, research and application, attract technical talents and innovative teams to move in, and provide intellectual support for the construction of technology.

For cities with insufficient financial resources, it is recommended to adopt the technological–environmental mode. Such cities should actively pay attention to the policy orientation of higher levels, take the initiative to strengthen communication and docking with higher government departments, provide timely feedback on difficulties and problems encountered by the city in the process of smart construction, and strive for more attention and support from higher levels in the formulation of policies and the allocation of resources; actively explore regional cooperation mechanisms, and build and share infrastructures with neighboring developed cities, such as jointly constructing regional cloud computing centers and data platforms, and jointly engage in the procurement of services to reduce construction and procurement costs by virtue of scale advantages; innovative PPP models or franchising methods can also be explored to attract social capital to participate in the construction of smart cities. There are often differences in the innate conditions for each city to build a smart city. Local governments should analyze the strengths and weaknesses of local conditions, build on their strengths and complement their weaknesses, choose a driving model that meets their own conditions, focus resources on promoting the development of key factors, and reduce the waste of overlapping resources.

As a complex systematic project, the promotion of smart city construction relies on the coupling of multiple factors, and there are significant differences in the dominant driving factors of different cities. While previous studies have emphasized the independent influence of a single factor on the high-efficiency construction of smart city pilots, this study discusses the multiple concurrent causal relationships affecting the high effectiveness of smart city construction through the fuzzy set qualitative comparative analysis method and reveals the synergistic mechanism of the configurational effects of the combination of multiple factors on the effectiveness of smart city construction. This study innovatively proposes that a single factor cannot constitute the necessary condition for the construction of a smart city, and that there is dynamic substitutability among factors. This finding breaks through the limitation of “single-factor determinism” in traditional research and provides a more flexible choice of construction paths for resource-constrained cities. Further, this study’s finding that financial resources, superior pressure, and technological human capital play key roles in high construction effectiveness is consistent with previous studies that emphasize institutional environment and resource dependence, but the configuration analysis further reveals their central roles in different combinations of paths.

Overall, this study achieves a paradigm shift from “single-factor analysis” to “multi-factor grouping” at the theoretical level, and systematically identifies three core driving modes (organizational, organizational–environmental, and technological–environmental) and their specific combinations of factors. The study not only identifies potential substitution relationships among factors, but also enriches the explanatory power of the TOE framework in complex public governance contexts. The findings clearly demonstrate the asymmetric characteristics of high construction efficiency and non-high construction efficiency paths, a finding that emphasizes the possible essential differences in the driving logics of smart city construction success and failure and deepens the understanding of the complexity of the construction process. At the practical level, this study constructs differentiated construction paths based on the differences in urban resources, which provides an operable basis for the selection of strategies for organization-led, resource-driven and technology-led cities, and also provides a new analytical framework and empirical evidence for the academic research and practical exploration in the field of smart city construction from a configurational perspective.

Finally, it should be noted that this paper also has certain limitations: due to the constraints of sample size, fsQCA also limits the number of conditional variables. This study selects seven conditional variables based on the TOE framework, but there are other variables not included in the discussion. Although the existing conditional variables are selected based on extensive literature, there may still be some subjectivity. Therefore, future research could explore different conditional variables, construct different analytical frameworks, and more comprehensively examine the factors influencing the high construction efficiency of a smart city pilot. Additionally, methods such as necessary condition analysis (NCA) could be used to analyze the necessity of influencing factors further and determine the degree of necessity.