Introduction

With the rapid advancement of science and technology, globalization of the economy, and advent of the post-industrial era, global economic development faces an escalating array of challenges. The frequent occurrence of natural environmental disasters, public accidents, and community safety events has significantly impacted urban governance. Artificial Intelligence, the Internet, ICT technologies, cloud computing and other digital technology-centric domains continue to evolve1, offering innovative perspectives, tools, and approaches for urban governance. Since the 1990s, with the initial introduction of the “smart city” concept, emphasis has been placed on the role of emerging information and communication technologies (ICT) in enhancing the urban infrastructure within cities2. In 2008, IBM introduced the concept of “Smart Planet”, positing smart city development as a pivotal breakthrough for addressing the governance challenges posed by the expanding size of cities.

Thus far, smart cities have remained a nebulous concept, with the deep integration of digital technology infusing them with new dimensions in their development3. The essence and definition of smart cities are transitioning from abstract concepts to tangible practices, evolving from singular functions to more complex systems4.

As China’s urbanization process accelerates, the urban population continues to expand, leading to increasingly prominent urban challenges, including environmental pollution, energy scarcity, and traffic congestion. Consequently, traditional urban governance models are no longer viable for adapting to contemporary economic landscapes. Since 2010, the Chinese government has issued a multitude of policies and directives, establishing the construction of smart cities as a pivotal pathway for advancing economic and social development. As of 2017, the Ministry of Housing and Urban-Rural Development of China had released three rounds of smart city pilot programs, rigorously exploring the scientific methodologies for the construction, management, and development of smart cities, thereby substantially enhancing the level of urban intelligence in the process of construction and development. A report from the Internet Data Center in January 2023 indicated that China is expanding its smart city initiative, with government-led investments in the smart city Information and Communication Technology market reaching 21.4 billion in 2022, a 21% increase from 2021. In contrast to the conventional one-dimensional governance model, which relies heavily on state power, smart cities emphasize the active engagement of diverse social organizations in urban governance. Departing from the traditional governance approach that primarily resorts to stabilization, smart city construction not only transcends the constraints of “strong administrative dominance”, but also broadens the scope of multi-party participation, fosters synergistic development in smart governance5, and enhances the precision and efficiency of urban management. However, the development of smart cities in China still lacks a long-term institutional framework. Most cities remain at the conceptual stage of smart city construction, facing a multitude of issues, including inefficient resource allocation, inadequate cooperation, and the absence of information-sharing mechanisms6. These problems significantly impede the advancement and achievement of modernized urban governance capacity and systems.

Smart city collaborative governance, regarded as comprehensive system engineering, entails open and diverse stakeholder participation in decision-making and execution across various sectors, hierarchical levels, and regions7. It is characterized by perceptual innovation, integration, and collaboration8, and is marked by features such as multidisciplinary cross-boundary and cross-___domain information sharing, robust business interconnections, extensive coordination demands9, intricate relational networks, and advanced technology4. Following the progress of the new generation of information science, there has been a marked increase in the number of network users. Extensive network access services, varied information dissemination channels, comprehensive online public services, and real-time online interactions on government-citizen platforms enable enterprises, social organizations, and the public to easily access firsthand and instant information. This enables prompt feedback on their needs and encourages active participation in the governance process, transforming them from passive subjects of governance to engaged participants. Interconnections among diverse entities, including both human and non-human elements, have become increasingly complex, giving rise to a sophisticated network of relationships10. The multiplicity of governance subjects has led to intricate interrelationships, posing a challenge for current smart city governance to effectively facilitate communication, overcome cross-border barriers in multiple fields, and unify all societal levels, various organizational entities, and diverse resources for collaborative governance. In response to these challenges, modern information and network technologies have played a critical role in creating an interconnected platform for information exchange. This process involves eliminating bottlenecks, streamlining the channels for information transmission, and fostering links between governmental governance bodies for the sharing of information resources. Consequently, a new pattern of collaborative governance is emerging: one spearheaded by the government, interconnected with various departments, bolstered by enterprises, and actively engaged by society. Smart city collaborative governance represents a vast system with complex dynamics resulting from the interaction of numerous internal and external factors, which necessitates the construction of a scientific, rigorous, and universally applicable theoretical model.

Previous studies have investigated the construction and governance of smart cities, focusing primarily on the dimensions of data-driven thinking and technological empowerment. From a data-driven perspective, successful smart city development requires extensive participation from citizens, society, and various organizations in construction and governance processes11. This requires collaboration and communication among diverse entities to achieve unified coordination and responsive actions4, thereby generating a substantial amount of data and information. Consequently, data governance has emerged as a central component of smart city governance12. From the perspective of technological empowerment, the existing literature highlights the significance of ICT in the construction of key infrastructures, such as post and telecommunications, transportation, and others. The expansion of the scope of ICT application is instrumental in enhancing governance efficiency13. The evolution of digital technology has led to the emergence of a data-based open governance network that presents a novel model for smart city management14. This model not only significantly enhances the efficiency of public services but also improves overall city operations15, enabling dynamic monitoring of the urban environment and providing real-time data feedback to ensure the effective operation of the city10. However, some scholars remain skeptical regarding the construction effectiveness of smart cities. They contend that while smart city development is undeniably reliant on technical support, an excessive focus on smart cities may obscure the disruptive impact of technological uncertainty on urban security16. Smart cities exhibit a high degree of dependency on the Internet and information technology, rendering them susceptible to becoming targets of hackers, thereby exposing urban network information security to increased risk17. In the context of smart city construction, governments procure technology and operational services from private enterprises, thereby enabling these enterprises to influence public decisions through their offerings. However, this arrangement may potentially undermine the public values and interests of the city18.

In summary, the current research primarily focuses on the impact of factors related to data-driven thinking and technological empowerment on smart city governance, and few studies have explored the combined effectiveness of multiple factors. This phenomenon frequently results in excessive dependence of smart cities on a solitary factor, impeding the optimal utilization of resources through rational allocation. In-depth research remains scarce regarding the attainment of multi-element collaborative governance and the elucidation of the operational and refined governance mechanisms within smart city collaborative governance. From a system perspective, this study aims to establish a smart city collaborative governance system, analyze the composition of the elements within the system and their interrelationships, construct a system dynamics model, elucidate the operational principles of the smart city collaborative governance system, and provide valuable insights for the development of a theoretical and methodological system for smart city governance and policy formulation.

Literature review

The concept of “smart city” is regarded as an effective model for addressing urban development challenges and has garnered widespread adoption by governments worldwide. Collaborative governance in smart cities has emerged as a focal point within the academic community.

Smart city governance

The notion of a smart city epitomizes an evolutionary concept that follows the information city and digital city, signifying a higher echelon of urban information technology19. Presently, smart cities, which have advanced technological capabilities, have transitioned from traditional to innovative approaches, offering substantial support for the sustainable and rapid development of metropolises and representing a theoretical frontier that requires constant exploration in urban research20. Smart cities are intrinsically linked to government management, and governments are tasked with ensuring the development of smart cities and prioritizing the development of certain regions21,22. The concept of smart city governance lacks a clear definition, and Zhang et al.23have categorized it into four levels: technology, economic development models, urban governance methods, and new governance mechanisms. Letaifa5 proposed the concept of smart governance from the perspective of the governance framework of smart city management, which involves the government’s combination of technological rationality and governance value concepts, and the construction of a holistic and sustainable smart development and governance system. Washburn et al.24 contend that the establishment of smart cities can bolster the efficiency and functioning of cities by effectively integrating, synergizing, and sharing resources across traditional infrastructure, education, healthcare, security, housing, transportation, and other social sectors. Tan et al.25 pointed out that achieving technology-driven smart cities necessitates simultaneous reforms in socioeconomic factors, human resources, legal systems, and regulations. Meijer et al.26 emphasized that smart governance is not solely about technology, but rather a complex process of institutional change. In this challenging information age, Zhang et al.27recognized that the realization of sustainable urban development transcends traditional technologies, requiring the seizing of opportunities to accelerate technological innovation, as well as the establishment of more advanced intelligent systems to achieve greater social benefits. With the profound integration of the IoT and urban development, smart cities have increasingly become a new focus of global urban development28. During the development of smart cities, governmental departments often excessively focus on technological, digital, and economic dimensions while neglecting social impacts, including social justice, equity, and inclusion. This prioritization has led to a multitude of issues including data security and privacy concerns, information silos, data barriers, and ambiguous subject positioning29. Luca et al.30 also noted that, in the current phase of China’s urban development, governing smart cities is essential for promoting the transition between old and new forms of urbanization, driving innovative regional development, and fostering the comprehensive construction of a moderately prosperous society.

Smart city collaborative governance

Collaborative governance, also referred to as cooperative governance, underscores the importance of diversifying governance subjects and standardizing systems. This shift signifies a transition from managing singular entities to multi-stakeholder governance31. Collaborative governance merges multi-center governance theory with the principles of collaboration, which include an emphasis on governance autonomy and the diversity of entities found in multi-center governance theory, as well as ideas related to cooperation, coordination, and order from collaborative studies32. Collaborative governance necessitates multiple governance entities to fully leverage the strengths and potential of all parties involved, foster efficient system operations, stimulate synergistic effects, and establish an interactive mechanism that ensures orderly functionality33. A city constitutes a complex, dynamic, coupled, and adaptive network system that continuously evolves and adapts to new environments. According to the complex systems theory, the growing complexity of the system necessitates an increasingly urgent need for coordination, yielding more significant synergistic effects34. New digital technology serves as a pivotal engine for enhancing the efficiency of urban governance, and “smartness” has emerged as the latest approach to propel the advancement of urban modernization to a superior level. With the progressive deepening of smart city governance reforms coupled with the intricate challenges confronted by cities and the involvement of diverse stakeholders, multi-agent collaborative governance has emerged as a significant direction in the reform of urban smart governance35. Collaborative governance underscores the importance of multiple stakeholders, including governments, social organizations, and citizens, fully leveraging their respective roles in establishing an efficient and orderly governance network. This network aims to optimize resource allocation and enhance governance effectiveness, thereby providing rational assurance for contemporary urban governance36. Despite notable advancements in the construction of smart cities from theoretical frameworks to practical implementations, significant practical challenges remain, including the absence of a co-governance concept, inadequate procedural connectivity, and inappropriate resource allocation37. These issues are poised to exacerbate the functional differentiation among various subsystems, such as the economy, politics, and law, thereby enhancing the complexity of urban societies38. Smart urban governance constitutes a network of actions that encompasses diverse stakeholders including governments, businesses, and city residents. The multifaceted nature of urban subjects, behavioral heterogeneity, and unpredictability of pivotal interactions collectively contribute to the formation of a sophisticated urban system39. Within the framework of smart city collaborative governance, focus should be placed on data-driven thinking, technological empowerment, and implementing mechanisms for risk communication, behavioral oversight, and stakeholder interest consideration to ensure equitable collaboration and improve governance efficiency40. Smart city collaborative governance involves citizens and stakeholders actively participating in the process of urban governance, thus inherently imbuing urban governance with a public nature41. Divergent governance concepts and goals among governments, enterprises, and social organizations can lead to varied approaches to cooperation, where each governance entity establishes its information platform, often operating independently, which fosters “closed-door” practices. This leads to a trust deficit and proliferation of “information silos” without cultivating a shared sense of collaboration42. Smart city governance represents a process and mechanism characterized by efficient interactions among diverse stakeholders. It constitutes a form of urban governance centered on citizen needs, driven by technological advancements, and facilitated by multi-subject participation and cross-departmental collaboration43. He44suggested that to maximize the effectiveness of collaborative smart city governance, it is essential to provide equal participation opportunities for each entity in smart city governance. Stakeholders in smart city governance assume distinct roles and responsibilities, and their interactive processes catalyze improvements in urban governance, economic growth, and environmental protection45.

Participatory governance is a pivotal form of urban governance in the forthcoming era. The development of smart cities necessitates a transition from a government-dominated management model to a collaborative governance framework that involves multiple entities, including government-led organizations, enterprises, and scientific research institutions. Government departments must not only improve internal management but also strengthen synergy with other stakeholders to collectively address urban challenges. Cooperation formed by multiple entities through profound technological integration and enhanced digital capabilities has emerged as an effective governance method for realizing public value46,47. Dvir et al.48emphasized that governments ought to create an environment in which citizens can exercise their subjective agency, harness their talents, and foster their desire to communicate and innovate. Batagan49 advocates for smart governance as a system in which government departments collaborate through communication to promote economic growth and genuinely serve the public. Kourtit et al.50 view smart governance as a proactive and open governance system in which all involved entities can effectively enhance their social and ecological effectiveness, respond to negative externalities, and overcome historical path dependency. Feng et al.51 proposed that smart governance necessitates adherence to innovative, integrated, green ecological, open and shared development framework, making full use of the advantages of modern information technology resources, with the aid of experiential, networking, data-driven, intelligence and other technological means, to promote the boundary regulation of public services and the innovation of the methods of “network + administrative public service”. This results in the construction of a modern urban governance model characterized by resource complementarity, multi-stakeholder collaboration, seamless government-citizen interaction, accurate public service, and intelligent public management.

Review

The surge in smart cities originates from the imperative to address the efficiency and sustainability challenges in urban areas52. In a complex societal context, the construction of smart cities requires the coordination of diverse subsystems within the urban system while accounting for the expression of varied interests, thereby facilitating the transformation and enhancement of urban governance. Smart City Collaborative Governance represents an emerging urban management model founded on information technology solutions, technological risk prevention, and various technological applications that promote urban development through a more comprehensive and targeted approach. In recent years, an increasing number of scholars have explored the collaborative governance of smart cities from various perspectives, transitioning the focus from singular urban governance models to multistakeholder collaborative governance models. Despite this shift, prior research has primarily been analyzed from governmental and relevant departmental perspectives, failing to systematically explore collaborative governance among multiple entities in smart cities. Furthermore, these studies have not elucidated the governance mechanisms underlying the roles of the diverse stakeholders within this context. Indeed, the complexity of modern urban societies has become increasingly evident, rendering it challenging for a single entity to manage governance risks arising within cities. Consequently, there is a pressing need to explore the governance potential of diverse entities and to establish a diversified and collaborative governance framework. This study contributes to several dimensions: First, based on the system perspective, it analyzes the structural complexity of the smart city collaborative governance system from the subject, object, and environment subsystems, scrutinizing the composition of elements at different levels, and clarifying the interactions among them. Second, the study constructs the smart city collaborative governance system based on systems thinking, delving into the interactions among system elements and formulating a smart city collaborative governance system dynamics model that includes the subject, object, and environment subsystems. This approach provides theoretical insights into the inherent mechanisms of smart city collaborative governance and theoretical support for the formulation of policies in this ___domain. Third, this article considers Dongguan City as an example to simulate future development trends of collaborative governance in smart cities under various policy models, adjusting policy parameters to identify and evaluate policy outcomes. This simulation-based approach provides practical guidance in shaping collaborative governance policies in smart cities.

City selection

Dongguan City is situated in the central part of Guangdong Province adjacent to Hong Kong and Macao. Geographically, it is part of the economic zone between Guangzhou and Shenzhen and plays a pivotal role in the Guangdong-Hong Kong-Macao Greater Bay Area and the Guangzhou-Shenzhen-Hong Kong-Macao Science and Technology Innovation Belt. By the end of 2022, the city was comprised of 32 towns (streets), encompassing 350 villages and 246 communities, and boasted a registered population of 10.437 million, including 9.6281 million urban residents, with an urbanization rate of 92.25%. Since the reform and opening-up, Dongguan, occupying less than three-thousandths of the nation’s land area, has generated more than one-hundredth of the national industrial output value, standing as a remarkable epitome of reform and opening-up in Guangdong and across the nation, and is listed among the 18 typical areas for national reform and opening-up. By the end of 2022, the local gross domestic product increased at an average annual rate of 6.0%. The average local gross domestic product per capita exceeded 100,000 RMB, indicating a high-income economy. The proportion of the three industries is 0.3:58.2:41.5. It has been recognized as a new first-tier city for four consecutive years and holds the rank of 17th in the “Comprehensive Economic Competitiveness Index of Chinese Cities”. Dongguan City emphasizes its positioning in “technology innovation + advanced manufacturing”. The number of sizable industrial enterprises surpassed 13,800, securing the title of a national model city for strong quality. Emerging industries, including new generations of information technology, high-end equipment manufacturing, new materials, new energy, artificial intelligence, and biomedicine, are rapidly developing. Seven strategic emerging industry bases have been established.

Since 2018, Dongguan City has actively coordinated the advancement of digital governance reforms and the construction, put in place a comprehensive management mechanism for digital government, advanced the construction of “cloud-network-digit” digital infrastructure, implemented a government data brain, amassed over three billion pieces of data, and broadened digital applications in fields such as medicine, education, social security, transportation, emergency response, urban management, and regulation.By enhancing government service efficiency and optimizing the business environment in terms of service delivery models and quality, Dongguan City continuously improves the satisfaction and sense of achievement of enterprises and the public. In the “China Urban Digital Governance Report (2022)”, Dongguan’s digital governance index is ranked 17th nationwide; the city won the “Government Service Reform Innovation Award,” “Best Practice Award for Government Service,” and “Service Innovation Award for Government Service” in the National Government Service Evaluation in 2020. In the Asia-Pacific Smart City Selection, Dongguan was awarded the title of “2021 China Leading Smart City,” and the “YiMaTongCheng” electronic citizen card received an Excellent Case Award in the Pioneer List of Smart Cities. At the 10th and 11th Global Smart City Congress, Dongguan received the “Resilience Innovation Award” and “Governance Award” for World Smart Cities. Therefore, in this study, Dongguan City was selected as the sample because of its representativeness.

Model construction

System structure

From a systems theory perspective, collaborative governance in the smart city constitutes a complex system comprising various factors. The subsystems within this system exhibit complex interactions and constraints with the collaborative governance process emerging through their interplay53. The collaborative governance system of a smart city primarily encompasses three subsystems: subject, object, and environment. These subsystems operate independently, yet interact with each other. By reinforcing the multi-agent coordination mechanism, accurately identifying governance targets, and optimizing external support, the system achieves seamless coordination and dynamic adjustment among subsystems. Consequently, this enhances the efficiency of smart city collaborative governance54. Specifically, the subject subsystem relies on the external support and constraints provided by the environment subsystem and exerts its influence on the object subsystem through decision-making and execution processes. The object subsystem, serving as the governance target, provides feedback on governance effectiveness to the subject subsystem, while its performance is constrained by the resource conditions of the environment subsystem. The environment subsystem supports the operations of both the subject and object subsystems by enhancing the institutional environment and improving information technology. Furthermore, changes within subject and object subsystems can influence the environment subsystem. The system structure framework is illustrated in Fig. 1.

Fig. 1
figure 1

Framework diagram of the system structure.

Full size image

The subject subsystem

In the smart city collaborative governance system, the subject subsystem holds a dominant position, encompassing diverse entities, such as government agencies, businesses, social organizations, and the citizenry55. The evolution and development of the entire system relies on the participation and management of various subjects to achieve the effective operation of the collaborative governance system of smart cities56. Therefore, the subject subsystem’s influence on smart city collaborative governance is primarily manifested through the scale and participation of each key stakeholder, categorized into four dimensions: governmental management capacity, social organizational capability, enterprise service capacity, and human resource capability. Specifically, these dimensions encompass sub-factors such as the scale of government financial allocations, transparency of governmental affairs, size and engagement of social organizations, digital operational proficiency of enterprises, proportion of highly skilled personnel, and citizen participation.

The object subsystem

The object subsystem of smart city collaborative governance comprises a comprehensive system that includes economic regulation, public services, market supervision, social management, and environmental protection, among others. The governance object constitutes the primary objective that must be achieved by the subjects involved in smart city collaborative governance. Each subject meticulously identified and effectively managed this object through the orchestration of material, energy, and information flows, thereby facilitating the realization of intelligent and refined urban management57. Therefore, the effectiveness of the object subsystem is primarily evident in its governance scope and the performance of its elements, including the level of smart economic development, the sustainable development capacity of the smart economy, the coverage and effectiveness of smart public services, smart ecological construction, and the effectiveness of environmental protection, etc. Specifically, the object subsystem comprises various sub-factors, including urban GDP, the level of smart community coverage, income of urban and rural residents, business income from digital industrialization, energy consumption per unit of output value, per capita green area in parks, and other pertinent indicators.

The environment subsystem

The environment subsystem serves as the external framework that supports and constrains the operation of the smart city collaborative governance system, facilitating material flow, information exchange, and energy distribution, among other subsystems. The environment subsystem is chiefly reflected in the comprehensive nature of the environment for smart city collaborative governance, encompassing the maturity of institutional and legal frameworks, network infrastructure construction, information technology support capabilities, and cultural environment development58. Specifically, the environment subsystem encompasses sub-factors, including total urban population, laws and regulations related to collaborative governance, ecological environment index, comprehensive website performance indicators, openness of data publication, and the number of base stations.

Study boundary and basic assumption

Study boundary

The system boundary comprises both spatial and temporal boundaries. In this study, the spatial boundary of the system was confined to Dongguan City, whereas the temporal boundary spanned 2015 to 2030. Historical data encompassing the period from 2015 to 2021 served to ascertain pertinent parameters and refine the model, providing a realistic benchmark for verifying the accuracy and validity of the simulation model. The period from 2022 to 2030 is dedicated to model simulation and scenario prediction, enabling the simulation and forecasting of potential development trends and evolutionary pathways of the system in the future as well as facilitating corresponding policy analysis.

Basic assumption

To effectively streamline the intricate collaborative governance system of smart cities in practical scenarios, this study introduces the following fundamental assumptions: (1) Factors external to the model are assumed to have no impact on the system. (2) During the simulation timeframe, the system was presumed to remain largely unchanged, with the development of its constituent elements maintaining relative stability, and the underlying trends remained consistent.

System causal loop diagram

The causal feedback loop of the system was carefully delineated to reflect the interrelationships among variables, as depicted in Fig. 2. In constructing the causal loop diagram, this study primarily considers three levels: Firstly, it examines the causal relationship between the subject and object subsystems. Specifically, enhancements in the government’s intelligent service capabilities, scale of social organizations, and quality of citizens can lead to improved social governance effects and increased public service coverage. Conversely, effective public services and social governance can foster greater citizen participation and satisfaction, thereby promoting optimization of government services59. Secondly, this study examined the causal relationship between the object and environment subsystems. The advancement of intelligent public services and environmental supervision hinges on the support provided by information technology and refinement of laws and regulations. Furthermore, the optimization of the institutional environment and the enhancement of public service quality stimulate the innovation and application of information technology, ultimately facilitating the realization of intelligent social management60. Thirdly, this study explores the causal relationship between environment and subject subsystems. Improvements in information technology and the institutional environment can enhance the quality of citizens, bolster the government’s intelligent service capabilities, and augment the innovation capacity of enterprises. Additionally, advancements in the government’s management ability and increased enterprise participation will foster the continuous enhancement and optimization of the institutional environment61.

Fig. 2
figure 2

Causal feedback loop of smart city collaborative governance system.

Full size image

System flow diagram

By integrating the causal feedback loops established in the previous section regarding the smart city collaborative governance system and the relationships between various variables and factors within the system, equations delineating the mathematical relationships among the selected state variables, auxiliary variables, and rate variables were formulated to establish the dynamic model of the smart city collaborative governance system. The system flow diagram was constructed using the Vensim-PLE software, as illustrated in Fig. 3.

Fig. 3
figure 3

System flow diagram.

Full size image

Data source

The model selected seven state variables, including citizen participation, total population, website comprehensive performance, GDP, ecological environment index, energy consumption, and output value of the information and communication industry. Additionally, it incorporates seven rate variables: improvement in citizen participation, population growth, enhancement in website comprehensive performance, GDP increase, improvement in the ecological environment index, energy increase, and output value increment of the information and communication industry. The auxiliary variables considered were business income, per capita GDP, financial expenditure on education, and so on. The primary data for the variables were obtained from the “Dongguan City National Economic and Social Development Statistical Bulletin”, “Dongguan Statistical Yearbook”, and “Guangdong Province Statistical Yearbook” for the years 2015–2021. The values of the variable parameters were obtained from the initial value, expert estimation, table function, and entropy. For variables that cannot be directly quantified, estimations are based on historical data. For instance, data such as the number of collaborative governance laws and regulations were computed using the arithmetic mean method, whereas estimates for variables such as the elimination rate of safety hazards, proportion of disaster prevention and emergency management expenditure, and number of base stations and social management demonstration applications were determined using linear regression or development trend analysis. The modelling process integrates Dongguan City’s data with variables undergoing continuous refinement to more accurately reflect the interrelationships among them. This process culminates in the formulation of the dynamic equation for Dongguan City’s smart city collaborative governance system.

Model simulation

Model test

As a complex system, the representation of smart city collaborative governance systems in terms of actual system effectiveness necessitates validation of the constructed dynamic model. The results of the model adaptability tests are as follows:

Adaptability test

The adaptability test primarily involves an intuitive and operational examination of the the smart city collaborative governance dynamic model. Initially, we assessed whether the variables and feedback loops in the subject, object, and environment subsystems accurately represented the actual system and whether the settings of the state and rate variables aligned with the objectives of the study. Subsequently, the Vensim software was employed to verify the structural integrity and dimensional consistency of the model. The constructed dynamic model closely aligns with the actual system structure and passes the dimensional consistency test.

Historical test

The historical test entails a comparison between the simulated results of the model and historical data. Typically, an error margin within 10% indicates a reasonably high level of fit between the constructed model and historical data. One state variable from each of the three subsystems of smart city collaborative governance was selected for the historical test, which examined the simulated values of the enterprise’s main business income, per capita GDP, and per capita public library holdings against the actual values, to assess the model’s fit. The error values from the test were below 10%, as shown in Table 1. This suggests a strong level of fit in the constructed model for smart city collaborative governance, demonstrating a robust predictive capability.

Table 1 Results of the historical tests.
Full size table

Scenario model settings

Policy simulation is a vital function of system dynamics, modifying specific policy variables within the system model to analyze their impact on the system output. By simulating different policy scenarios, a range of operational results for the smart city collaborative governance system can be illustrated, offering a basis for determining the optimal scenario.

Variable selection

The proportion of high-tech personnel and scale of social organizations were selected as policy adjustment variables for the subject subsystem. These variables were chosen to investigate the impact of enterprise and social organization behavior on the smart city collaborative governance system. From the environment subsystem, policy adjustment variables including e-government disclosure systems, the completion rate of municipal and county public legal service centers (which include legal aid convenience windows), the percentage of investment in basic network infrastructure, and the per capita expenditure on education, culture, and entertainment were selected. These variables are designed to specifically examine the impacts of institutional, legal, technological, and cultural environments on the smart city collaborative governance system. E-government disclosure systems and the completion rates of municipal and county public legal service centers (which include legal aid convenience windows) are largely determined by government departments. Therefore, as policy adjustment variables, they can also serve to examine the influence of government departments on the collaborative governance systems of smart cities. The per capita expenditure on education, culture, and entertainment primarily reflects citizens’ spending behavior regarding education, culture, and entertainment, making it an effective policy adjustment variable for assessing the impact of citizen behavior on the smart city collaborative governance system.

Policy model setting

To examine the effects of policies on the smart city collaborative governance system and assess the impacts of different scenario models, six scenario models were designed to simulate the smart city collaborative governance system in Dongguan. The values of the policy adjustment variables in the different models were primarily determined based on the actual investment values in Dongguan from 2015 to 2021, along with comprehensive considerations in comparison to other cities. The specific analyses are as follows:

The average proportion of high-tech personnel in Dongguan is approximately 18.28%. Guangzhou, adjacent to Dongguan, exhibited similarities in industrial development; however, its value for this indicator was approximately 30% higher than that of Dongguan during the same period. Beijing’s proportion of high-tech personnel was notably higher, at approximately 80% above Dongguan’s value during the same period. It becomes clear that Dongguan possesses significant potential for improvement in the development of high-tech industries and in attracting high-tech personnel. Considering the differences in Dongguan’s foundational development and urban functionality relative to Guangzhou and Beijing, expectations for this indicator’s future setting include an increase of 10–30%.

The scale of social organizations in Dongguan is approximately 3.729. When compared with neighboring cities, such as Guangzhou and Shenzhen, they exhibit values approximately 10% and 40% higher, respectively, during the same period. Nationally, the scale of social organizations surpasses that of Dongguan by approximately 35%. This indicates that Dongguan has a significant potential for growth on the scale of social organizations. Given the limited scope of Dongguan’s urban environment, a substantial increase in the scale of social organizations may be challenging to achieve in the short term. Consequently, the projection for this indicator is an anticipated increase of 10–30%.

Dongguan’s e-government disclosure score is approximately 42.011, while Hangzhou, which is a model for government information disclosure in China, exhibits a value roughly 6% higher during the same period. This demonstration shows that Dongguan’s government information disclosure system is among the higest in the country. However, when compared with a typical city, there remains a noticeable gap. Hence, the targeted increase for this indicator was set at 2–6%.

The completion rate of public legal service centers (including legal assistance windows) in Dongguan’s cities and counties was approximately 10.914; in Hangzhou, this rate was approximately 40% higher during the same period. Considering the differences in urban scale between Dongguan and Hangzhou, the projected setting for this indicator is an expected rise of 10–30%.

The proportion of investment in basic network infrastructure in Dongguan was approximately 1.359. For this indicator, Guangzhou’s value was approximately 45% higher, and Hangzhou’s value was approximately 85% higher during the same period. This indicates that Dongguan must accelerate smart city development and boost investment in the basic network infrastructure. Consequently, the target for this indicator is projected to increase from 20 to 40%.

The per capita expenditure on education, culture, and entertainment in Dongguan is approximately 3700.857 RMB. During the same period, Hangzhou’s value for this indicator was approximately 12% higher. Therefore, the target for this indicator is set to increase by 5–15% in the future.

In conclusion, six prospective development models for the smart city collaborative governance system in Dongguan were conceptualized, as illustrated in Table 2. The detailed analyses are presented below:

(1)Basic Model: This model pertains to the existing policy inputs for all adjustment variables within Dongguan’s smart city collaborative governance system. The system’s operational outcomes have been examined in prior simulation forecasts, and this model serves as a baseline for comparison.

(2)Stable Development Model: Policy adjustment variables in the basic mode undergo moderate increments, signifying a balanced augmentation in policy inputs for Dongguan’s smart city collaborative governance system, targeting stable progression throughout the system.

(3)Rapid Development Model: In this framework, there is a sustained and escalated investment in pertinent policy adjustment variables to catalyze swift development within Dongguan’s smart city collaborative governance system. Hence, the adjustment variables were further elevated beyond the parameters of the stable development model.

(4)Technology-Led Model: This paradigm represents an intensified future investment in Dongguan’s smart city infrastructure, notably in foundational network systems and high-tech personnel, striving to propel the robust development of the smart city collaborative governance system through augmented technological investments. As a result, the variables corresponding to the proportion of high-tech personnel and network infrastructure investment were calibrated to be 30% and 40% higher than the basic model, respectively, while other policy adjustment variables were incremented to a median level between the stable and rapid development models.

(5)Social-Led Model: This approach highlights the amplified role of social organizations and citizens in Dongguan’s smart city collaborative governance system, utilizing societal dynamics to spearhead system advancement. Therefore, this approach stipulates that the variables for the social organization scale and per capita expenditure on education, culture, and entertainment are augmented by 30% and 15% higher than the basic model, respectively, with a modest enhancement from the stable development model for other policy adjustment variables.

(6)Government-Led Model: This construct centers on the prospective development of Dongguan’s smart city, predicated on augmented inputs from governmental entities into institutional and legal frameworks to facilitate collaborative governance, with such departments propelling the positive evolution of the smart city collaborative governance system. Thus, this construct adjusts the variables for the government information disclosure system and completion rate of public legal service centers in urban and county areas to be elevated by 6% and 30% higher than the basic model, respectively, with other policy adjustment variables being raised to an intermediate level between the stable and rapid development models.

Table 2 Proportion of improvement of situational pattern variables.
Full size table

Results analysis

Using the Vensim software, simulations were carried out for the six scenario models previously established to understand the outcomes of the simulated variables within the Dongguan’s smart city collaborative governance system, as depicted in Figs. 4, 5, 6, 7, 8 and 9. The specific analyses are as follows:

Firstly, in the Basic Model, observed variable simulations within Dongguan’s smart city collaborative governance system generally trended towards the lowest or highest levels. Among these variables, energy consumption per unit of output was the highest across the six models, while the values of the other five observed variables remained lower than those in the other five scenario models. It is apparent that under the Basic Model, the overall development level of Dongguan’s smart city collaborative governance system is relatively low, necessitating the exploration of optimal inputs for various policy adjustment variables within the system to foster its development.

Secondly, under the Stable Development Model, the observed variable simulations showed improvements compared with the Basic Model. Increasing the proportion of high-tech personnel by 10% and the infrastructure investment proportion by 20% created favorable conditions for the transformation from traditional industries to high-tech and digital industries, markedly boosting digitized industrial revenue compared to the Basic Model. This increase led to a notable reduction in energy consumption per unit output in traditional industries, which was facilitated by improvements in digital infrastructure, thereby greatly improving the overall ecological index62. Concurrently, smart infrastructure upgrades, enabled by investment in basic facilities and the development of digital industries, considerably improved the system’s operational efficiency, greatly advancing the collaborative governance of Dongguan’s smart city. Furthermore, a 2% increase in the transparency of government functions significantly improves the comprehensive performance of government websites. The increase in the completion rate of public legal service centers, along with the enhancement of residents’ cultural quality, contributed to increased civic engagement in the collaborative governance of the smart city. Thus, even minor enhancements in the policy adjustment variables of the smart city collaborative governance system could expedite material, energy, and information flow within the system, substantially boosting the operational effectiveness of Dongguan’s smart city collaborative governance system.

Thirdly, under the Rapid Development Model, the observed variable simulations exhibited significant improvements compared with the Basic Model. The continuous increase in the proportion of high-tech personnel and infrastructure investment by 10% each from the Stable Development Model drove the digitized industrial revenue to approximately 8% higher, reaching an expected value of 2,330.53 billion RMB by 2030. This model resulted in a substantial reduction in energy consumption per unit of output compared with the Basic Model, significantly improving the ecological environment index, with an expected 7% difference between the two by 2030. An increase in the transparency of government functions and the completion rate of public legal service centers contributed to a slight improvement in smart community coverage and the comprehensive performance of government websites compared to the Stable Development Model. By elevating the scale of social organizations and per capita expenditures on education, culture, and entertainment by 20% and 10%, respectively, the capacity of citizens to participate in the smart city’s collaborative governance was significantly enhanced. Overall, enhanced inputs into the policy adjustment variables of the smart city collaborative governance system, along with adjusting the ratio of policy inputs and expanding demand-driven policy inputs, can foster enthusiasm for social participation in the smart city, leveraging market competition, positive feedback incentives, reduced participation costs, and diverse social engagement to promote smart city construction63. This accelerates the internal operation of the system and significantly improves the operational effectiveness of Dongguan’s smart city collaborative governance system.

Fourthly, under the Technology-Led Model, the observed variable simulations surpassed those of the three previous models. The most striking feature of this model is the substantial increase in the proportion of high-tech personnel and infrastructure investment, the largest of all models evaluated. Among the six models, the direct impact of this model yielded the highest output values for smart community coverage and digital industrial revenue, with an anticipated reach of 2,489.3 billion RMB by 2030. Although the other four policy adjustment variables saw only minor enhancements from the Stable Development Model, the combined effects of these adjustments enabled this model to excel in projecting future trends for energy consumption per unit output and the ecological environment index. Energy consumption per unit output was the most efficient among all models, reaching a low value of 0.226. Digital technology has revolutionized the collection speed, analytical efficiency, and dissemination mechanisms of various societal and data resources in urban governance, offering diverse entities increased pathways for governance participation and action62. This results in efficient digital tools and technological methods for collaborative governance, further improving smart community coverage and thus encouraging multi-agent governance. Improvements in the technological environment also ensured the construction and enhancement of government websites, even with fewer institutional and legal environment inputs than the Fast Development Model. Therefore, the comprehensive impact resulted in an even more significant improvement in government website performance compared to the Fast Development Model. Although the increases in social organization size and per capita expenditure on education, culture, and entertainment were more modest than those in the Fast Development Model, the enhanced application of basic network infrastructure and digital information technology broadly improved the city’s informatization and intelligence levels. This supports the government in building online e-government platforms, realizing e-government activities, and advancing digitization, thereby enhancing the speed and accuracy of information dissemination, facilitating effective resource sharing among various departments, minimizing redundancy, noise, and interference in information, and amplifying the administrative efficiency and decision-making of public service departments64. This made it easier for citizens to participate in collaborative governance in smart cities. Consequently, the increase in citizen participation in this model is marginally higher than that in the Rapid Development Model. Thus, the Technology-Led Model has a significant impact on the enhancement of the smart city’s collaborative governance system and represents an effective strategy for future development of Dongguan’s smart city.

Fifthly, under the Social-Led Model, the observed variable simulations demonstrated a significant improvement. With a substantial increase in the scale of social organizations and residents’ per capita expenditure on education, culture, and entertainment, the citizens’ capacity to participate in the smart city’s collaborative governance reached a pinnacle among the six models, with an anticipated growth of approximately 24.354% by 2030 compared to the Basic Model. Despite a slight decline in digitized industrial revenue compared with the Rapid Development Model, there was a marginal increase in the Stable Development Model, resulting in significant improvements in smart community coverage and energy consumption per unit of output. The rise in civic engagement has significantly influenced the monitoring and protection of the urban environment. Grass-roots citizen groups can be highly valuable in urban governance; relying on citizens not only facilitates the formation of practical problem-solving programs, but also enables a supervisory and evaluative role65. Coupled with the development of the urban digital industry and reduction in energy consumption, the combined effect is projected to elevate the urban ecological environment index to the highest level among all models. The government’s investment in institutional and legal environments has increased by 3% and 15%, respectively, and the expansion in the scale of social organizations coupled with active citizen participation has collectively contributed to raising the comprehensive performance of government websites to a higher level, with an expected increase of approximately 1.808% by 2030 compared to the Basic Model. In summary, the Social-Led Model exerts a weaker impact on the city’s digital industrialization development, but significantly influences citizen participation, the ecological environment, smart community coverage, and government website performance, rendering it an effective option for the future development of Dongguan’s smart city.

Finally, the observed variable simulations were consistently higher in the Government-Led Model. In this model, a 15% increase in the scale of social organizations and an 8% increase in per capita expenditure on education, culture, and entertainment considerably enhance citizens’ active participation in smart city collaborative governance. The government’s increased investment in institutional, legal, and technological environments has elevated the comprehensive performance of government websites to the highest level among the six models, with an expectation of reaching 97.759 by 2030. The cumulative effect of these factors resulted in a higher level of citizen participation in the collaborative governance system than in the first four models, ranking slightly below the corresponding value of the Social-Led Mode. Digital industrial revenue was only slightly lower than the corresponding value of the Technology-Led Model. The extent of the increase in the proportion of high-tech personnel and infrastructure investment surpassed that of the Stable Development Mode, yielding improvements in smart community coverage and energy consumption per unit of output that surpassed those of the Stable Development Model. Increased citizen participation has directed greater attention towards the city’s ecological environment. Combined with the city’s digital industrial development and reduced energy consumption, these impacts have a positive influence on the city’s ecological environment, ranking second in each model’s simulated values, only below the Social-Led Mode. Consequently, the Government-Led Model demonstrates a moderately positive effect on improving the smart city’s collaborative governance system, significantly impacting industrial digitalization and ecological environment enhancement, and considerably boosting the comprehensive performance of government websites, establishing it as a viable option for future development of Dongguan’s smart city.

Fig. 4
figure 4

Simulation curve of the comprehensive performance scenario model of the government websites.

Full size image
Fig. 5
figure 5

Simulation curve of the citizen participation scenario pattern.

Full size image
Fig. 6
figure 6

Simulation curve of scenario model of smart community.

Full size image
Fig. 7
figure 7

Simulation curve of the ecological environment index scenario pattern.

Full size image
Fig. 8
figure 8

Simulation curve of digital industrialization.

Full size image
Fig. 9
figure 9

Simulation curve of energy consumption scenario model per unit of output value.

Full size image

Discussion

Smart city collaborative governance functions as a complex dynamic system in which the interrelations among its internal elements govern the collaborative governance process. This study develops a smart city collaborative governance system using a systemic approach; establishes a system dynamic model based on the interaction among the internal elements of the subject, object, and environment subsystems; and conducts policy simulations in Dongguan City to investigate the varying impacts of different policy implementation intensities on smart city collaborative governance.

Smart city collaborative governance comprises of numerous elements. Zhai et al.46 and Wang et al.47 argue that smart city collaborative governance requires diverse entities to establish cooperation to achieve effective governance. They also underscored the leading role of the government in the construction of smart cities and the need to establish stable and strong collaborative relationships between the government and other participating entities. Lu et al.66 highlighted the close connection between urban intelligence, governance entities, city development, and the pursuit of multiple entities. However, they did not delineate other elements required for collaborative governance beyond diverse entities. This research originates from a systemic perspective and develops a smart city collaborative governance system, concentrating on the theoretical framework and logical analysis of the subject, object, and environment subsystems of collaborative governance. It details the composition of system elements and their interrelationships, thereby broadening the theoretical and methodological framework for research on smart city collaborative governance.

The intricate operational mechanisms within the smart city collaborative governance system dictate the collaborative governance process. Song67argued that urban collaborative governance requires comprehensive, multidimensional, multilevel, and multistage effective collaboration across all governing bodies to foster the fusion and symbiotic cooperation of internal and external resources. Advocating for the active participation of social organizations in collaborative governance, the promotion of collaborative engagement between various organizations and governments, and diverse interactive responses68, results in governance and social affairs that are more significant than the sum of their parts69. However, these studies have not comprehensively delineated how to achieve efficient collaboration and collaborative governance mechanisms among multiple entities in smart cities. In light of this, our study utilizes system dynamics to develop a dynamic model of the smart city collaborative governance system. This model quantifies the degree of mutual impact between the various involved elements within and among the subject, object, and environment subsystems of collaborative governance, effectively uncovering the intrinsic operational mechanisms of the smart city collaborative governance system, expanding the quantitative analysis methods for smart government collaborative governance mechanisms, and enriching research on the smart city collaborative governance path.

Collaborative governance in smart cities has markedly improved the effectiveness of governance. Zhang et al.70observed that collaborative governance represents an advanced form of governance and that effective collaboration raises the efficacy of collaborative governance. In governance practice, various participating entities experience initial cognitive acceptance, mid-term cooperative interactions, and subsequent stages of collaborative co-governance, progressively advancing at each level, which significantly contributes to enhancing the effectiveness of public cultural governance71. Prior research has largely investigated the methods and approaches of urban collaborative governance, yet it lacks precise metrics to measure the effects of enhancing urban collaborative governance effectiveness, rendering the offered policy suggestions less effective in providing scientific guidance for practice. Based on the system dynamics model, this study constructed six scenario models to simulate the dynamic trend of the collaborative governance system of the smart city in Dongguan under different policy patterns. This model adjusts the parameters of different elements within various subsystems to examine the effects of changes in the intensity of element inputs on smart city collaborative governance, thereby offering theoretical and practical guidance for the formulation of smart city collaborative governance policies.

Conclusion

This study utilized data from Dongguan City from 2015 to 2021 and applied system dynamics methods to simulate and model smart city collaborative governance to provide scientific guidance for smart city policy formulation.

In the baseline model, where no adjustments are made to the regulating variables within the smart city collaborative governance system and future investments continue as per current policies, the observed trend in system variables is the poorest among all the models. This indicates a relatively low overall development level of Dongguan City’s smart city collaborative governance system, emphasizing the necessity for increased policy variable investments to achieve the beneficial development of the collaborative governance system.

Increased policy input into participating entities and governance environments to enhance the operation of the smart city collaborative governance system promotes system development. The intensity of the policy input directly influences the system operational efficiency, demonstrating that the simulation of a rapid development model with stronger policy input outperforms the stable development model.

The policy-regulating variable input induces collaborative effects among various subsystems within the smart city collaborative governance system. The proper allocation of policy input enables advantageous complementarity among policy-regulating variables, resulting in optimal policy input benefits. The impacts of technology-led, social-led, and government-led models on Dongguan City’s smart city collaborative governance system posses distinct focuses, jointly advancing the system’s benign operation.

Based on these conclusions, the following recommendations are proposed:

The government should strengthen its overall coordination, concentrate on key areas, adopt a long-term perspective, and engage in scientific planning. Considering that smart city development is a complex, systemic project that requires comprehensive consideration of various factors for resource allocation, establishing a strong coordinating body is essential. Therefore, in the initial construction phase, government-led efforts are crucial for unified planning, gradual progress, and rational and effective allocation of resources, technology, and personnel72. The developmental needs of smart city collaborative governance dictate that the government and other participant networks must be closely integrated for effective collaborative governance. As an “invisible hand”, the goveronment should firmly control the three key aspects of smart city development: planning, construction, and governance. It should increase the intensity of policy input, regulate and eliminate “blockages” and “nodes” in the network of relationships between various actors, break down cross-barriers, and avoid the “island effect”. The government must thoroughly consider the behavioral disparities among diverse stakeholders, foster the profound integration of digital technologies, novel management paradigms, and traditional urban governance elements, augment the collaborative governance capacities of multiple stakeholders, and cultivate shared values to spearhead the reform of collaborative governance mechanisms.

The government should design robust systems, augment financial investment, and advance the application of cutting-edge digital technology. Utilizing big data to capture and analyze the diverse emotional and demand differences among various city entities fosters technology-driven human-resource interaction, thereby forming an effective feedback mechanism for service demand15. For instance, AI technology that captures real-time information changes among multiple entities can parse instant data, mine fine-grained knowledge, extract features, and perform variable calculations and simulation comparisons, thereby providing insightful suggestions for policy formulation. Consequently, the government needs to expedite the in-depth integration of AI technology with urban governance, inject new impetus into digital transformation, and foster a faster and higher-quality development of smart city construction73. Furthermore, the utilization of digital technology to establish “data warehouses” enables the timely processing of real-time feedback data, thus activating the value of data. Through access to instant information, the efficiency and accuracy of government decision making can be enhanced, policies can be strategically invested in key areas, and networked linkage effects can be unleashed, yielding maximum returns with minimal investment. In the context of collaborative governance within smart cities, it is equally crucial to reinforce data security management and personal privacy protection, mitigating the risks of data leakage and misuse while safeguarding the legitimate rights and interests of citizens. The government should strive to continually enhance training programs and incentive mechanisms for scientific and technological talents, bolstering cooperation among governmental bodies, enterprises, universities, and scientific research institutions. Additionally, it is imperative to promote technological innovation and the commercialization of research achievements, thereby providing intellectual and expertise support for the advancement of smart city construction.

The government should adhere to the principle of “people-oriented”, improving the system of citizen participation, ensuring reasonable access to government information, and expanding information disclosure. Only citizens with sufficient knowledge of government information can provide effective feedback on specific issues. Reducing participation costs guarantees the rights of entities to information and participation, fostering full engagement in collaborative governance and contributing to the development of intelligent solutions. Utilizing the internet and digital technology to develop a unified platform and information system broadens the avenues for democratic participation and stimulates public engagement in city collaborative governance. Furthermore, information transfer channels between departments will be opened to facilitate vertical information interoperability and data sharing between superiors and subordinates, and horizontally to dismantle data barriers and information silos, thereby enhancing the level of governance and achieving synergy between functions and data. Collaborative governance in smart cities necessitates the government to fully exercise its guiding and coordinating roles, while augmenting the availability of market and social resources. The government should harness the strengths of social organizations in the realms of urban management, social environment, and environmental protection. It is essential to establish open and transparent information-sharing platforms, broaden avenues for disseminating information to the public, and strengthen citizens’ sense of participation and identification with urban governance.

Simulation of the smart city collaborative governance system constitutes a dynamic process. Only through the continuous adjustment of the model parameters in response to the real system’s operations and policy changes can the model remain effective. This paper presents certain limitations, including the exclusion of minor influential factors when delineating system model boundaries for simplicity’s sake; Owing to the unavailability of some data, certain exogenous variables were subjectively set, potentially influencing the simulation results of the model. Future research will focus on further refining and enhancing the model.