Conceptually Related Smart Cities Services from the Perspectives of Governance and Sociotechnical Systems in Europe

Abstract

The absence of a comprehensive smart city governance model has prompted research into the characteristics of the relationships among cities, services, and stakeholders. This study aims to identify, from the perspectives of governance and sociotechnical systems, the characteristics of conceptually related smart city service implementations based on stakeholder partnerships. Social network analysis was utilized based on existing research datasets. Stakeholders, services, and representative European sustainable smart cities were included in the dataset in relation to this study’s operational definition. The first finding is that the initial conceptually related smart city services are reflected in the accumulated and current characteristics of the smart city services. These depend on stakeholder partnerships, while the network foundation differs between the initial and latter services. The second finding clarifies how different development services depend on stakeholder partnerships and how multiple stakeholders, including local entities, are vital to deal with current challenges in massive urbanizations. The third finding demonstrates the emerging roles of private sectors and some intermediate services in the global network of cities. This study contributes to the management of smart cities by identifying how service development occurs based on stakeholder partnerships and contributes to their theoretical basis by empirically demonstrating the importance of multi-stakeholder partnerships to address current urbanization challenges.

1. Introduction

Sustainable smart cities are complex living ecosystems that involve diverse stakeholders participation. Both public and private sectors have led urban development that mobilizes information communication technologies as it requires substantial funds and infrastructure. The technologies are geared toward developing several smart city capabilities to satisfy users’ demands, so it becomes essential to consider the citizens’ participation and to satisfy local needs in the context of socio-technical transitions [1]. These trends make cities powerful places to observe and pilot urban transformation [2,3,4]. The United Nations (2022) emphasizes the digital development agenda in cooperation with diverse stakeholder partnerships, particularly engaging citizens and private sectors in a whole-of-government and whole-of-society approach globally [4]. According to the UN-Habitat World Cities Report 2020, smart cities are rapidly deploying technology to address various challenges and to meet the digital development agenda [5] by collaboration across multiple fields, including urban planning, transport planning, administration, healthcare, economics, infrastructure, environment, weather, safety, security, public services, community engagement, and research and innovation [4]. Smart cities have evolved from technological platforms for managing urban resources to innovation generators with the participation of public, private, citizen and non-governmental sectors to satisfy local demands and deal with local challenges [6]. In other words, smart cities are not only innovative engines or platforms to pilot technologies and provide services, but also solutions in the form of living ecosystems to challenges that arise when collaborating with diverse stakeholders [7,8]. The subsystems or independent services that are mediated by data and ICTs are interactive and interrelated in sociotechnical transitions [9].

There is a lack of appropriate smart city governance models based on holistic approaches to the relationships between dual top-down and bottom-up stakeholder systems and service systems. There is a discourse that smart cities are complex Systems-of-Sub-Systems integrating interactive systems between ICT-based urban services and diverse stakeholders [10,11,12]. Many researchers have suggested smart city governance frameworks that consolidate hard infrastructure, including technologies, infrastructure, services, and data agencies, and recently even emphasize citizens’ participation [13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31]. On the other hand, there is an attempt to identify the sophisticated elements of the governance model from bottom-up and top-down approaches using systematic literature reviews or surveys [8,32,33]. However, there is a lack of a comprehensive governance model dealing with service evolution based on technologies and investments with different stakeholder partnerships from perspectives of sociotechnical transitions. It is particularly challenging to combine top-down policies with bottom-up endeavors [12].

The absence of smart city governance models results in a variety of definitions or visions of sustainable smart cities in theoretical and political domains. This results in the various names applied to smart cities and an unconditional frame of either technology-oriented or citizen-participatory initiatives. In addition, empirical research has been conducted in specific fields or areas [34]. However, smart cities involve many communities, and very large amounts of capital are required to build human settlements that consider multiple scales, and physical and soft infrastructure that includes telecommunication, water, energy, waste, built environments, diverse urban services, knowledge, and regulation through the participation of diverse stakeholders. Moreover, smart cities are interoperable with other information systems, which is different from traditional cities [35].

The challenges of developing a smart city governance model lead to the comprehensive research question: “How are the stakeholders’ partnership systems networked with the conceptually related smart cities’ services implementation from perspectives of governance and sociotechnical systems?” As it is difficult to identify the relations among services, stakeholders, and smart cities, three subordinate research questions are raised: (1) What are the characteristics of services in the evolution of conceptually related smart cities? (2) Which services of the conceptually related smart cities are developed depending on different stakeholder partnerships? (3) Which services and stakeholders are mostly connected, assuming conceptually related smart cities are connected virtually and physically?

To address these research questions concerning current issues in smart city governance, this study aims, from the perspectives of governance and sociotechnical systems, to identify complex characteristics of governance models for conceptually related smart city services implementations that depend on stakeholder partnerships with the projection of developing a new research area. This research covers three objectives: (1) to clarify the characteristics of services in the evolution of conceptually related smart cities by expanding the existing literature, (2) to demonstrate how different phases of the conceptually related smart cities services are developed depending on different stakeholder partnerships, and (3) to identify connected services and stakeholders, assuming that conceptually related smart cities are connected virtually and physically.

This paper is structured with six sections. Section 2 reviews the concept of conceptually related smart cities and establishes a theoretical background on smart cities’ complex systems through a literature review. Section 3 describes the framework of the social network analysis with data. Section 4 describes three findings regarding the three objectives. Section 5 discusses and expands our findings using existing theories, and the last section presents concluding remarks.

2. Literature Review

2.1. Concepts of Conceptually Related Smart Cities

Conceptually related smart cities fall under the umbrella of sustainable smart cities. The concept of sustainable smart cities synthesizes two distinct concepts, sustainability, and smart cities. Both of these concepts have unique characteristics that complement each other’s weaknesses. [31]. The sustainability indicators, which focus on social, economic, environmental, and political aspects, contribute to smart cities’ human and socio-technical perspectives. Conversely, smart cities can contribute to compelling technical aspects of sustainability and vice versa [13,18,36,37,38]. The existing literature defines sustainable smart cities with five branches that include: (1) a smart city with a vison of the sustainable development goals or sustainability [39,40,41,42,43,44,45], (2) ICT-embedded cities for managing resources [46,47,48], (3) innovative platforms for achieving urban sustainability [40,49,50,51,52], (4) transformative processes based on evolutionary concepts of ICT and multi-stakeholder partnerships [53,54,55,56], and (5) innovative sociotechnical transitions in digital transformation in correlative networks between multi-stakeholders and various services to satisfy local needs [31,34,37,48,57]. In the last concept, partnerships with diverse stakeholders results in niche innovation or technological trajectories which influence the sociotechnical regime, including industries, policy, laws, and urban services that transform the urban landscape in a digital transformation history [58]. In particular, Kim and Yang (2021) defined sustainable smart cities as complex systems to develop and implement diverse urban services from social, technological, governance, environmental, and management aspects in sustainable and integrative manners [31]. Mora et al. (2021) bring new theoretical insights by adopting a middle-range of the transitions and examine the theoretical discourse based on empirical testing in the view of the multilevel perspective [48]. Calzada (2017) emphasizes a multilevel governance devolution scheme in the transition of smart city initiatives by analyzing four European cities from the preservice of techno-politics systems [59]. By extending Calzada’s (2017) idea of devolution in multistakeholderism, Kim and Yang’s (2021) definitions, and Mora et al.’s (2021) perspectives based on the last concept, this study defines sustainable smart cities as transformative initiatives for developing and implementing integrative social, technological, governance, economic, environmental, and management urban services sustainably.

The conceptually related smart cities that have been developed in the sustainable smart city paradigm are diverse and have evolved in cooperation with various stakeholders through the use of information and communication technology (ICT). The cities have evolved in the five stages of IT transformative history with the cooperation of the public, private sectors, citizens, NGOs, and academia through mobilizing ICT. The stages opened with the invention of pre-telecommunications, the telegraph, telephony, computing, and ICT [60,61,62]. The cities have advanced the co-creation value by exchanging knowledge and sharing data between citizens and the city. [63]. The cities are considered evolutionary forms based on society and technologies [15,31,34,64,65,66,67], while they are organized as distinctive forms alternate to smart cities [46,68]. As this study expands the transformative concept of sustainable smart cites with diverse stakeholders’ participation, the conceptually related smart cities are also characterized by diverse names such as cyber cities, virtual cities, Internet cities, wireless broadband cities, intelligent cities, digital cities, ubiquitous cities, and smart cities. [34]. These cities have evolutionary traits in socio-technical transition and can be viewed as ICT-based transformative communities and cities. Overall, these conceptually related smart cities reflect a holistic approach to sustainable smart city development involving diverse stakeholders’ participation.

2.2. Smart Cities as Complex Systems

The concept of cities as complex systems has been developed by various fields, including social science, ecology, business management, and smart city studies, after it was introduced in the biological sciences fields in the early 20th century [12]. It was later adapted and expanded by Ludwig von Bertalanffy through his revised General Systems Theory [69]. When the concept initially appeared in contemporary science, it ignored the interactions of various fields and emphasized isomorphic laws to unify individual sciences vertically as an organized wholeness. Bertalanffy argued that the existing theory would ignore local events and dynamic interaction manifest in mathematical approaches and suggested that in open systems approaches, the entropy of the system, through dynamic interconnection, becomes a fixed arrangement in the models of equivalent, feedback, and adaptive behavior. The modified general systems theory transformed the concept of cities into distinct collections of interacting entities in equilibrium, which firstly influences the planning and management process in top-down approaches, such as in ideal cities, for example, Ville Radieuse, or those of other modern architects [70]. Michael Batty (2017) explained that the spatial structure is in an equilibrium stage even though technologies and fashions have triggered many social, economic, and environmental changes and that cities are in states far from equilibrium in consideration of urban dynamics in historically evolved economic cycles which coincide with scientific and technological advances, cultural movement and migrations of the population in relation to climate changes or physical conditions [71]. In addition, the physical equilibrium is described as out-of-sync with disequilibrium events, even if they are subsystems in cities [72]. The perspectives of scales in the urban environment create diverse stances toward the phenomenon of development of cities. Cities have unique systems when it comes to micro perspectives, as Michael Batty mentioned, while from macro perspectives, they have universal and spatiotemporal systems. The urban system dynamics, which Jay Forrester demonstrates, dealt with the life-cycle of city development based on three internal structures of the model, consisting of three subsystems, namely industry, housing, and people, which are controlled by an external environment [73]. The model proposed that adjustment between the attraction of internal systems and the total attractiveness of the city needs to develop the accommodation capacity of the city [73]. Bettencourt and West (2010) found that the scale of cities is a significant determinant of the characteristics of cities, and that the development route follows the city size, which positively correlates to crime rates, GDP, and income, so the scale of cities is called a scaling law. [74]. They also showed that urbanization makes cities greener, more efficient, more prosperous, and safer as the cities adapt [74]. On top of the scaling law, Bettencourt et al. (2007) discovered successive cycles of super linear innovation, which are led initially by the biological organization, by sociotechnical organization in the middle, and later by an individual organization, and these processes are tied up with the degree of urbanization, economic development, and knowledge creation [75]. In other words, the time scales of emerging innovation become shorter as the population increases and is more connected than before. Smart cities comprise multiple intelligently connected systems and integrate material construction, agencies, cultures, living creatures, and services in systems of sub-systems.

Smart cities are intricate systems comprised of subsystems that integrate interactive systems between ICT-based urban services and a diverse range of stakeholders. Within the smart city service model, terms such as factors, services, domains, components, systems, layers, and sensors are often used to refer to various segments, while groups of segments are commonly referred to as architecture, models, systems, frameworks, and dimensions. To address the interoperability issues between technologies and services and to facilitate knowledge expression methods or maximize synergies between ICT infrastructures, various telecommunications and electronic field researchers have introduced service-oriented reference architectures from technological perspectives [63,76,77,78]. The other perspectives prioritize multi-stakeholder partnerships and co-creation networks by exchanging knowledge [63,79,80,81,82,83]. Numerous researchers have also organized service models from complex sociological, environmental, and economic perspectives to establish indicators that analyze city data and rank the cities [13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31]. Some studies have suggested service models to identify the evolutionary paths of service development in the history of smart cities and intelligent communities [15,65]. The ICT-based services and multi-stakeholder partnerships facilitate the transformation of e-government to smart governance [84]. City governance, which the government develops, is an extended-term with multi-faceted and multilevel systems of stakeholders, sectors, and agencies [33,84].

The development of smart cities has led to the transformation of traditional e-government into smart governance. The concept of smart city governance has been extensively discussed in the context of socio-technical transitions. The theory from multilevel perspectives was first framed by Geel [58], who was influenced by Nelson-Winter’s evolutionary economics [85], Dosi’s technological trajectories [86], Malerba’s Sectoral systems [87], Carlsson’s technological systems [88], and Bijker’s technological systems in relation with sociology, institutions, and rules [89]. Smart city governance has been developed from e-government and digital government. Unlike the government, which refers to a form of authority, governance refers to a process of action, processes, traditions, and institutions by which collective decisions are made and implemented [84]. In the context of digital transformation, e-government focuses on implementing digitizing technologies to formulate data based on existing analog government, while digital government emphasizes satisfying local needs and users’ demands by re-engineering and re-designing services and processes [90]. In the streams of transformations, smart cities are researched from socio-technical perspectives to contribute to a theoretical smart city governance model. Mora et al. (2021) analyze the landscape of smart cities from socio-technical dynamics perspectives. The city is a complex and evolutionary adaptive system for urban innovation and urban sustainability in cooperation mainly with public and private sectors and a few with civil society [48]. Kim and Yang (2023) analyze the empirical characteristics of conceptually related smart cities services’ evolution from perspectives of socio-technical transitions based on multi-stakeholders and found different services advancement depending on phases of partnership and common characteristics of developing services regardless of stakeholders’ partnership [34]. Some researchers applied it to the governance model. Nam and Pardo (2011) connect urban governance to e-government and innovation to make cities smarter from a techno-political perspective [68]. Calzada (2017) researched the transition of four smart European cities regarding the techno-politics of data from the perspectives of multi-level governance devolution schemes [59]. Waart et al. (2016) emphasize the networking of top-down and bottom-up elements in transitions of smart city dynamics [91]. The existing literature reveals that the socio-technical transition consists of rich perspectives explaining smart cities’ dynamics based on technical and social analysis that develop a theoretical understanding of techno-governance [8,92].

There is a dearth of appropriate models for governing smart cities, which can be attributed to several factors, including diverse visions, inconsistent implementation, and oversimplified technological solutions [12,33,93]. Current models tend to focus more on the interplay between technologies, services, data, and buildings while neglecting the crucially connected role of stakeholders partnerships and urban contexts. As an attempt to deal with the issue, Robert et al. (2018) proposed a conceptual model for smart city governance based on 13 indicators encompassing components such as services, technologies, stakeholders, legislation, and structures, as well as contextual factors and outcomes. [33]. In this regard, the key players in driving sustainable innovation are connected and agglomerated communities, individuals, and organizations on the basis of frontier technologies. Calzada (2017) demonstrates the importance of devolution in smart city development to increase the ownership and the self-responsibility of investment in infrastructure and data [59]. Additionally, ownership of data and cities has blown up the debate on multi-stakeholder participation since citizens and all stakeholders can be seen as tiny chips along with artificial intelligences inside a giant system to collect and analyze data, as Harari (2016) argued [59,94]. However, smart city implementations in real projects still suffer from fragmentation due to variations in definitions, as well as the lack of a model that reflects the multidimensional operational nature of cities and the importance of multi-stakeholder partnerships [10,11,93]. These challenges are aggravated by the lack of a model reflecting urbanization contexts and multi-stakeholder partnerships considering the multidimensional operational nature of cities [12].

This study addresses the gaps in the existing literature by identifying the characteristics of stakeholder partnership systems and their relationship to the implementation of sustainable smart city services. By doing so, this study seeks to contribute to developing a more comprehensive smart city governance model that considers the role of multi-stakeholder partnerships in realizing sustainable urban development.

3. Materials and Methods

Social network analysis is primarily utilized to identify the study aim based on the published data in Kim and Yang’s (2023) research [34]. As demonstrated in Figure 1, the study commences with a research question motivated by smart city governance challenges, as explained in the preceding section. The primary research question is then broken down into three objective research questions directly corresponding to the study objectives under the study aim. In essence, this research aims to identify characteristics of services implementations depending on the stakeholders’ partnerships from the perspectives of governance and sociotechnical systems that are established by the major research question, i.e., how the stakeholders’ partnership systems are networked with the conceptually related smart cities services’ implementation from perspectives of governance and sociotechnical systems. The aim is achieved through three research questions, each directly linked to three objectives. The first research question aims to clarify the characteristics of services in the evolution of smart cities, while the second focuses on demonstrating the different services phases developed depending on stakeholder partnerships. The third research question identifies connected services and stakeholders assuming that smart cities are connected both virtually and physically.

The research question is closely linked with the study aim and the concept, which comprise measurable indicators that establish the study framework [34,95], as illustrated in Figure 1. The concept encompasses six aspects, namely, social, technological, governmental, economic, environmental, and managerial factors, and a single keyword, namely, the sustainability of urban services. These aspects and keywords are utilized to provide a background for analysis and select target cities. European cities were selected as the target cities using cluster sampling, from a population of 221 smart cities which were investigated by Smart City Tracker 1Q18 [96]. The study population was derived by combining three ranking lists of sustainable smart cities that represent the six aspects of the concept. These ranking lists are the United Nations-Habitat Global Urban Competitiveness Report (for social, economic, environmental, and managerial aspects) [97], the McKinsey Company’s Smart Cities: Digital Solutions for a More Livable Future (for technological aspect) [98], and the United Nations E-Government Survey 2020 (for governmental aspect) [84]. As shown in

Table 1. A process of sampling cities.

Study Population	First Screening Result	Second Screening Result	European Cities
Singapore, Tokyo, New York, London, San Francisco (Oakland), Paris, Hong Kong, Osaka, Los Angeles-Long, Beach-Santa Ana, Chicago, Barcelona, Moscow, Stockholm, Seoul, Munich, Stuttgart, Boston, Madrid, Shenzhen, Frankfurt am Main, Philadelphia, Toronto, Taipei, Houston, Miami, Berlin, Melbourne, Rome, Shanghai, Seattle, Manchester, Atlanta, San Jose, Cleveland, Sydney, Hiroshima, Birmingham, Beijing, Milan, Montreal, Dallas-Fort Worth, Buenos Aires, Vienna, Tel Aviv-Yafo, Denver-Aurora, Hamburg, Zurich, Nagoya, Baltimore, Kitakyushu-Fukuoka, Copenhagen, Hannover, Salt Lake City, San Diego, Perth, Washington D.C., Incheon, Suzhou, Raleigh, Kuala Lumpur, Vancouver, Amsterdam, Astana, Geneva, Brussels, Detroit, Guangzhou, Austin, Orlando, West Yorkshire, Cologne, Helsinki, Daejeon, Istanbul, Ulsan, Richmond, Valencia, Jerusalem, Columbus, Sao Paulo, Bridgeport Stamford, Phoenix-Mesa, Nanjing, Doha, Haifa, Antwerp, Hartford, Riyadh, Sapporo, Gwangju, Busan, Naples, Xiamen, Milwaukee, Glasgow, Adelaide, Dubai, Daegu, Santiago de Chile, Malaga, Athens, Wuxi, Dortmund, Louisville, Pretoria, Essen, Tianjin, Foshan, Taichung, Brisbane, Auckland, Dresden, Saint Petersburg, Virginia Beach, Calgary, Las Vegas, Bogota, Medina, Dongguan, Wuhan, Lima, Kaohsiung, Dusseldorf, Tampa-St., Petersburg, Belfast, Jedda, Worcester, Hangzhou, Lyon, New Haven, Leipzig, Dublin, Hamilton, Hague, Buffalo, Charlette, Liege, Zaragoza, Torino, Colorado Springs, Chengdu, Qingdao, Nashville-Davidson, Macao, Rio de Janeiro, San Antonio, Zhongshan, Minneapolis-Saint Paul, Sendai, Lisbon, Silo, Ningbo, Lille, Liverpool, Provo-Orem, Changzhou, Zhengzhou, Amman, Venice, Dammam, Rotterdam, Tainan, Changsha, Leicester, Tehran, San Juan, Providence, Shizuoka-Hamamatsu M.M.A., Verona, Johannesburg, Baton Rouge, Bangkok, New Orleans, Gold Coast, Ottawa-Gatineau, Bologna, Leon, Solfa, Indianapolis, Shenyang, Pittsburgh, Ogden, Florence, Kansas City, Budapest, Montevideo, Zhuhai, Honolulu, Barcelona-Puerto La Cruz, Oklahoma City, Dallin, Minsk, Porto, Mecca, Xi’an, Ahvaz, Hefei, Marseille-Aix-en Provence, San Francisco, Tallinn, Roma, São Paulo, Mexico City, Warsaw, Prague, Almaty	Singapore, New York, Stockholm, Seoul, Shanghai, Amsterdam, Helsinki, San Francisco, Chicago, Copenhagen, Barcelona, Melbourne, London, Tokyo, Paris, Moscow, Madrid, Toronto, Berlin, Bogota, Buenos Aires, Istanbul, Brussels, Dubai, Mexico City, Sydney, Johannesburg, Lisbon, Athens, Kuala Lumpur, Seattle, Austin, Vienna, Beijing, Shenzhen	Singapore, New York, San Francisco, Chicago, Barcelona, Melbourne, London, Tokyo, Berlin, Dubai, Mexico City, Seoul	Barcelona, London, Berlin
207 cities	36 cities	12 cities	3 cities

Adapted from Kim and Yang (2023) [34], published by MDPI, 2023.

, the study’s first selection resulted in 36 cities after removal of cities mentioned more than twice in the three ranking lists. The top 20 smart cities from the rank of smart cities performance published by Juniper Research were integrated with the first screening results to select high-performing sustainable smart cities in the second sampling, which resulted in 12 cities. Finally, European cities were chosen because they have been leaders in e-government development within The United Nations e-government development index (EGDI), which is an index of online services, telecommunications, and human capital since 2010 [4,99]. As this study aims to identify characteristics of conceptually related smart cities services implementations from perspectives of governance and sociotechnical transitions, the selected cities embody the critical elements of the concept of sustainable smart cities and stand out as pilot cities for building smart city governance within the paradigm of sociotechnical transitions. For this study, the target cities are Barcelona, London, and Berlin.

On the basis of research questions linked with study aims, this research selects the data regarding conceptually related smart cities for three target cities. The data are from Kim and Yang’s (2023) article [34], which coded and weighted datasets on three relevant cities’ projects and plans from 1969 to 2021. In a sophisticated analysis aligning with the study aim and the concept of sustainable smart cities, the researchers formulate a city-level dataset, which is configured with categories including year, data sources, stakeholders, services, converted number of stakeholders, the sum of the converted number of stakeholders, and converted weights of services [34]. The data categorized with events, services, year, and stakeholder data were weighted and coded using the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol. The weights are assigned to each year as “1” equally, and the weight was then divided by the number of participating stakeholders in that year to distribute the divided weights to each stakeholder [34]. This means that individual services are given different weights based on the number of implemented services and participating stakeholders in that year, thereby giving equal weight to services and stakeholders that are correspondingly aligned in the same year [34].

A Social Network Analysis is utilized for this research. This is a specific application of graph theory that originated from Euler’s mathematical investigation to represent social actors’ networks using points and social relations using lines. This approach is based on German sociology, where Georg Simmel and others emphasized the formal properties of social relations [100]. Alfred Vierkandt and Ledpold von Wiese adopted terminologies of node, edges, and connections that make up social network analysis, while Moreno provided the idea of a sociometric, a type dataset depicting a metric of sociogram [101]. Social network analysis has been utilized to identify corporate power and interlocking directorships [102]. The successive research has explored the power and influence of banks [103]. It is also utilized to analyze community structures with business networks [104]. It has also been utilized to identify and map knowledge flow between organizations since it was introduced to regional and innovation economics fields [105]. Ma (2023) breaks up the collaborative innovation network into spatial and topological networks using social network analysis [105]. Radulescu et al. (2020) deploy it to define critical competencies and human resources in the innovation network from the collaborative model of a smart city [106]. Mora et al. (2019) utilize the analysis to show the two-dimensional network of actors collaborating to enable smart city development in New York City in a quadruple-helix collaborative model of stakeholder engagement [48]. Sconavacca et al. (2020) explore the active areas in the evolution of smart city research using the method [107]. Kim and Yi (2018) analyze the coherence between national and local smart cities plans regarding the keywords and service elements [108]. The existing literature on the method has been developed to reveal the relations between social activities and organizations in the urban environment. At the same time, it highlights the interdisciplinary traits of smart cities research in regards to diverse aggregation and disaggregation of services, technologies, infrastructures, and stakeholders.

Social network analysis is utilized to confirm the data relations among systems of services, stakeholders, and cities in regard to the research aim. It mainly utilizes Gephi 0.9.2 software, open-source software, and interactive tools used for visualization and examination or assessment of various simple and complex networks and dynamic and hierarchical graphs [109]. Gephi 0.9.2 needs two datasets for social network analysis: a node dataset comprising a network of actors and an edge dataset consisting of a list of relations between actors. The node list consists of three columns: ‘id,’ ‘label,’ and ‘attribute’. The label column includes the names of all actors, including services and stakeholders. The ‘id’ is set up with constant numbers, which take the role of links between the node dataset and edge dataset. The edge list utilizes the stakeholders, services, and their weights. The dataset is transformed into an edge list containing four columns: source, target, type, and weight. Then a column of source and target, which were initially filled with words, needs to substitute with the corresponding ‘id’ referring to an identical ‘label’ in the node dataset. After completing the dataset, Gephi 0.9.2 is used to analyze the relationship between service and stakeholders by importing the node table first and the edge table appended to the previously opened node table. Among analysis functions such as average degree, average weighted degree, network diameter, graph density, HITS, modularity, PageRank, connected components, and others, average weighted degree, betweenness centrality, and eigenvector centrality are utilized in this study. The average weighted degree reflects the cumulative number of keyword connections between surrounding keywords to obtain the node and connection frequency. The eigenvector centrality indicates neighbors’ weighted centrality, which considers not only the number of connected nodes but also other nodes’ centrality as the concept of eigenvector centrality basically states that if each actor is connected to a neighbor with a high connection centrality index, the influence on the network (eigenvector centrality) is greater than if it is connected to a neighbor that does not. The betweenness centrality, introduced by Freeman, is the shortest path based on the enumeration of identifying critical actors in the network [110]. A node with higher betweenness centrality can penetrate the blocked or siloed information from fields to fields through the node so that it has the potential to rise in power [111,112]. The three network analyses are visualized in a preview tab, and the detailed data are identified in a data laboratory tab. Lastly, the interpretation process demonstrates the research questions concerning the network between cities and services at first, then second, the stakeholders and services, and last, the cities’ organic network systems in consideration of three factors.

4. Results

4.1. Characteristics of the Conceptually Related Smart Cities Services

The early smart city development features are reflected in accumulated characteristics of the conceptually related smart city services depending on stakeholder partnerships. These results support and expand the existing literature. This paper provides results for the identical phases of three cities’ partnerships which were qualitatively measured in the existing literature, Kim and Yang’s (2023) study [34]. As shown in

Table 2. Data of Three Cities’ Weighted Degree, Eigenvector Centrality, and Betweenness Centrality.

		Barcelona						London						Berlin
		WD	Rank of WD	EC	Rank of EC	BC	Rank of BC	WD	Rank of WD	EC	Rank of EC	BC	Rank of BC	WD	Rank of WD	EC	Rank of EC	BC	Rank of BC
Stakeholders	Public	53.641	1	1	1	90.719	1	23.69	1	1	1	142.3	1	9.5833	3	0.9588	2	78.75	1
	Private	8.5936	4	0.9167	2	40.88	2	13.09	3	0.4862	6	24.538	2	13.468	1	1	1	56.75	2
	People	3.8318	10	0.7833	3	18.945	3	0.32	13	0.3211	11	1.4929	11	2.1151	6	0.6365	5	1.25	9
	Academia	1.9229	13	0.7448	4	12.6	4	5.39	6	0.4702	7	4.5381	10	0	19	0	19	0	11
	NGO	0.1538	21	0.2849	16	1.8554	16	2.18	9	0.5417	2	17.127	3	1	8	0.2859	10	0.25	10
Services	Infrastructure	17.353	2	0.5432	5	6.8139	5	18.09	2	0.4883	3	12.234	4	12.147	2	0.6431	3	13.5	3
	Social	7.9701	5	0.5007	8	2.0264	9	4.9	7	0.4883	3	12.234	4	3.5754	5	0.6431	3	13.5	3
	Economy	9.877	3	0.5007	8	2.0264	9	6.43	4	0.3947	10	8.7301	7	5.8459	4	0.4389	7	4	6
	Data	4.9591	8	0.5432	5	6.8139	5	2.04	10	0.4325	8	7.7126	8	1.0635	7	0.5789	6	6	5
	Government	5.6171	7	0.5007	8	2.0264	9	6.4	5	0.4325	8	7.7126	8	0.8155	10	0.2165	14	0	11
	Education	0.9575	15	0.5007	8	2.0264	9	3.85	8	0.4883	3	12.234	4	0.375	12	0.2165	14	0	11
	Environment	3.045	12	0.5432	5	6.8139	5	0.24	16	0.1739	15	0	15	0.25	13	0.2165	14	0	11
	Health	0.619	18	0.5007	8	2.0264	9	0.17	17	0.2675	12	1.0476	12	0.1111	15	0.2224	11	0	11
	Architecture	4.6015	9	0.5007	8	2.0264	9	0.58	11	0.1739	15	0	15	0.0238	18	0.2165	14	0	11
	Transport	6.0872	6	0.5007	8	2.0264	9	0.07	18	0.1739	15	0	15	0.8624	9	0.4389	7	4	6
	Safety	0.8163	16	0.2798	17	0.256	18	0.44	12	0.2675	12	1.0476	12	0.1012	16	0.4389	7	4	6
	Tourism	0.4192	19	0.2798	17	0.256	18	0.07	18	0.0843	22	0	15	0.037	17	0.2224	11	0	11
	Standardization	0.0379	22	0.1465	21	0	21	0.04	20	0.1739	15	0	15	0.5	11	0.2165	14	0	11
	Energy	3.1438	11	0.3223	15	3.1488	8	0.04	20	0.1739	15	0	15	0	19	0	19	0	11
	Waste	1.7134	14	0.2798	17	0.256	18	0.02	22	0.1739	15	0	15	0	19	0	19	0	11
	Media and art	0.2301	20	0.2599	20	0.4573	17	0.27	15	0.1739	15	0	15	0.125	14	0.2224	11	0	11
	History	0.6952	17	0.1465	21	0	21	0.32	13	0.2675	12	1.0476	12	0	19	0	19	0	11

Note: Abbreviations: WD = Weighted Degree, EC = Eigenvector Centrality, and BC = Betweenness Centrality. Yellow highlights refer to the leading partnerships in each city. Green highlights infer to services within the rank top 10, and the red words in the green highlights indicate upgraded ranks in the two centralities compared with the ranks in weighted degree.

, it is revealed that for the three cities, the public and private sectors are mostly leading stakeholders. When one is selected as the leading stakeholder and others in the other stakeholders, the analysis reveals Barcelona in a public–people partnership, London in a public–academic–NGO partnership, and Berlin in a private–people partnership, as indicated by yellow shades in Table 1. Moreover, this study expands the existing studies by identifying the early development of smart city services in the current development of smart cities and the accumulated local contexts of developing smart city services. The services in the top 10 weighted degrees, highlighted with green in Table 1, are related to the early development of smart city services illustrated in Figure 2. For instance, Barcelona developed social, economic, architectural, governance, transportation, data, and infrastructure services in the beginning during the launch of the first stage of the @BCN Plan. The city was developed to regenerate cities based on citizens’ creative ideas and infrastructure advancement for building a global smart city model in a holistic and comprehensive city renewal approach [56,113]. The services developed at the beginning in Barcelona in the Knowledge-Based Urban Development Project, which was an initiated smart city project, have higher weights in the accumulated network results of the weighted degree. The London services in the high rank were also mainly developed at the beginning of smart cities’ evolutions until the launch of the government plan, Inclusion Through Innovation, except for some temporal services regarding health, history, and standardization. The high-ranked Berlin services in the weighted degree analysis are infrastructure, social, economy, data, governance, and transportation services developed from the beginning of smart cities before the implementation of Silicon Allee. The results demonstrate that understanding the context of smart city development is crucial in developing smart cities.

In the context of smart city development, the early services carry significant weight, while high levels of network mediation by various stakeholders characterize the later services that evolve in sociotechnical transitions. Social network analysis provides quantitative information and correlations to understand network variables [12]. According to

Table 3. Three European Cities Services’ Weighted Degrees, Eigenvector Centrality and Betweenness Centrality.

	Barcelona	London	Berlin
Weighted Degree
Eigenvector Centrality
Betweenness Centrality

, the services with higher degrees of centrality, including eigenvector centrality and betweenness centrality, receive improved ranks in weighted degrees among the three cities. Barcelona and Berlin, whose partnerships commonly involve citizens, exhibit strikingly similar results when the two centralities are compared in each city independently. The citizen partnerships promote the highly connected services that are sustained and strengthened, such as education, environment, and health for Barcelona and safety services for Berlin. In other words, citizens become human resources or agencies to connect services through their data and active participation. Specifically, the architecture service in Barcelona, which has modest weight by itself, is linked with various services through the public, people, academia, and private sectors, thereby becoming prominent in the eigenvector centrality. Conversely, the energy service in the same city, which has low weight and few links with stakeholders, was downgraded in the eigenvector centrality. Furthermore, data and knowledge play a role in transmitting information throughout sociotechnical transitions. The services’ two centralities commonly receive higher weights than weighted degrees when analyzed as a whole, indicating that explicit and tacit knowledge is transferred from one generation to the next by being embedded in the development of infrastructure, social, economic, environmental, and other fundamental services while basic services intrinsically develop data services.

The characteristics of conceptually related smart city services reflect the features of services evolution in socio-technical transitions. The phases of partnership in each city support the existing literature. The smart city implemented context influences current development in that the high-weighted services reflect early-developed services. The early developed services have high weighted services, which refers high frequency of implementation, and later developed services have high degrees of network mediated by stakeholders. Amid an evolution, some intermediated services are highlighted recently. This result is identified through weighted degree, eigenvector centrality, and betweenness centrality in social network analysis.

4.2. Services Developments Depending on Stakeholders’ Partnerships

The development of different smart city services depends on partnerships with various stakeholders. Social network analysis can provide information on decision framing and key actors, and it is a relatively quick and easy way to conduct research that encourages participation from diverse viewpoints and actors [12]. Sustainable smart cities have been planned and developed in cooperation with the public and private sectors, as shown in Table 2, while diverse services are connected and networked through multiple agents, as illustrated in Table 3. Although the weighted degree results of all partnerships primarily emphasize the implementation of infrastructure and economic services, this is not the only reason for smart city development, which should not be overlooked as simply a project developing the economy through giant investments in infrastructure and technology. The connected services through cities and services are diverse and dependent on partnerships, as analyzed in

Table 4. Services Development depending on Partnerships.

Label	Private–People				Public–People				Public–Academic–NGO				Private–Academic–NGO				Public–Private				Public–Private–People–Academic-NGO
	WD	Rank of WD	EC	Rank of EC	WD	Rank of WD	EC	Rank of EC	WD	Rank of WD	EC	Rank of EC	WD	Rank of WD	EC	Rank of EC	WD	Rank of WD	EC	Rank of EC	WD	Rank of WD	EC	Rank of EC
Barcelona	6.2	5	1	1	29	3	1	1	28	2	1	1	5.3	5	1	1	31	3	1	1	34	3	1	1
London	13	3	0.5	3	23	6	1	2	31	1	1	2	21	1	0.8	2	36	1	1	1	44	1	1	1
Berlin	15	2	0.7	2	11	8	0.7	16	9.6	4	0.7	3	14	3	0.7	3	22	4	0.9	3	25	4	0.9	3
Infrastructure	16	1	0.4	4	72	1	0.9	3	23	3	0.4	4	18	2	0.4	4	32	2	0.4	4	39	2	0.4	4
Economy	9.2	4	0.4	4	28	4	0.9	3	8.6	5	0.4	4	9.2	4	0.4	4	17	5	0.4	4	4.6	8	0.4	4
Social	4.6	6	0.4	4	27	5	0.9	3	7.2	7	0.4	4	4.6	6	0.4	4	8.8	7	0.4	4	12	6	0.4	4
Data	1.4	7	0.4	4	14	7	0.9	3	3.2	10	0.4	4	1.5	8	0.4	4	4.3	9	0.4	4	0.6	19	0.4	4
Government	1	8	0.3	13	35	2	0.9	3	8.6	6	0.4	4	1.4	9	0.3	11	9.1	6	0.4	4	18	5	0.4	4
Education	0.5	10	0.3	13	9.3	9	0.9	3	3.9	8	0.4	4	2.3	7	0.3	11	2.3	11	0.4	4	9.6	7	0.4	4
Safety	0.3	13	0.3	9	6.1	13	0.9	3	0.7	16	0.4	4	0.4	11	0.4	4	0.8	16	0.4	4	4.5	9	0.4	4
Environment	0.3	11	0.2	15	7.1	12	0.9	3	1.6	12	0.4	4	0.3	13	0.2	16	1.8	12	0.4	4	1	15	0.4	4
Transport	0.7	9	0.3	9	8.6	11	0.9	3	3.7	9	0.4	4	0.7	10	0.3	14	4.4	8	0.4	4	0.3	20	0.4	4
Architecture	0.2	17	0.2	15	9.3	10	0.9	3	2.9	11	0.4	4	0.1	18	0.2	16	3	10	0.4	4	1.9	12	0.4	4
Standardization	0	19	0.2	15	2.7	17	0.6	17	1.1	14	0.4	4	0.3	15	0.3	11	0.9	15	0.4	4	1.1	14	0.4	4
Health	0.3	14	0.3	9	3.2	16	0.8	13	0.3	19	0.3	15	0.3	12	0.4	4	0.4	19	0.4	4	4.4	10	0.4	4
Tourism	0.2	16	0.4	4	0.7	20	0.4	20	0.1	20	0.2	20	0.2	16	0.4	4	0.3	20	0.4	4	1.6	13	0.3	18
Media and art	0.3	15	0.3	9	5.4	14	0.8	13	0.4	18	0.3	15	0.1	17	0.3	14	0.6	17	0.4	4	0.9	16	0.3	18
Energy	0.3	12	0.2	15	4.4	15	0.7	15	1.3	13	0.3	15	0.3	14	0.2	16	1.6	13	0.3	18	3	11	0.4	4
Waste	0.1	18	0.2	15	1.3	18	0.5	18	0.8	15	0.3	15	0.1	20	0.2	16	0.9	14	0.3	18	0.7	17	0.4	4
History	0	20	0	20	0.8	19	0.5	19	0.7	17	0.3	15	0.1	19	0.1	20	0.6	18	0.3	18	0.7	18	0.3	18

Note: Abbreviations: WD = Weighted Degree, EC = Eigenvector Centrality, and BC = Betweenness Centrality. Green (weighted degree) and Blue (eigenvector centrality) highlights refer to services within the top five rankings.

. Public sectors sustain and enlarge the services connection based on the fundamental services, while private sectors connect with emerging services that are different from public sectors. The fundamental services are infrastructure, economic, social, and data services. The services led by the public and other stakeholders are accumulated and affixed to service development phases of less influential partnerships compared with public–people partnerships, public–academic–NGO partnerships, and public–people partnerships. For instance, the standardization service in public–academic–NGO is added to the high-ranked public–people partnership services. Health, tourism, and media services in public–private partnerships are affixed to those in private–academic–NGO. In other words, the public–people partnership is becoming a means to develop smart cities from a humanistic perspective compared with the prevailing public–private partnerships. Even though there are gaps between partnerships with public sectors, some services are mainly led by their partnerships, including governance, education, safety, environment, transportation, and architecture services. Meanwhile, partnerships with the private sector and other players disclose different services connection than the ones with the public sector, even though their implementation seems to occur in an ad hoc way. Representative instances include tourism services in private–people partnerships and health tourism services in private–academic–NGO partnerships compared with identical partnerships with public sectors.

Indeed, multi-stakeholder partnerships are crucial in addressing complex issues such as energy and waste management in urban areas. Table 4 in the last column demonstrates that multi-stakeholder partnerships are one of the alternative solutions to deal with recently emerged urbanization challenges regarding energy and waste issues. These two issues are related to Co2 emissions, which are reduced by changed behaviors in cities resulting from compact or walkable cities, e-mobility in combination with low-emission energy sources, and enhanced carbon uptake and storage using nature [114]. The two services’ production is mismatched with consumption regarding geographical aspects. The energy is produced outside cities and consumed inside them, while the waste issue follows the reverse pattern [115]. Technological advancements such as smart meters and smart bins can assist in identifying the average demands for households and reducing waste. In addition, urban planning can play a crucial role in developing services close to demand centers, leading to simpler networks and lower costs. The integration of energy and waste services through district heating systems can also bring benefits in terms of energy savings. Barcelona’s Solar Ordinance is an excellent example of a multi-stakeholder partnership promoting renewable energy use [115]. By encouraging the installation of thermal solar panels in buildings, the city has achieved significant energy savings. Such initiatives not only promote sustainability but also have positive economic impacts, including reduced energy costs and increased job opportunities in the renewable energy sector.

4.3. Networks of Conceptually Related Smart Cities

The networks of conceptually related smart cities provide ways to improve stakeholder partnerships and service implementation systems. Social network analysis can facilitate understanding of socio-institutional structures, actors, linkages, and approaches to enhance knowledge transfer, including tacit and explicit knowledge [12]. When the three cities are highly connected, services related to data, education, environment, health, media, and tourism receive higher emphasis in eigenvector centrality compared with a weighted degree, as presented in

Table 5. Systems of Cities Network.

	Weighted Degree	Rank of WD	Eigencentrality	Rank of EC
Private	30.756	3	1	1
Public	58.546	1	0.92	2
People	4.3643	11	0.8262	3
Academia	6.2567	7	0.6943	4
NGO	3.2436	12	0.5887	5
Infrastructure	38.396	2	0.524	6
Data	4.6273	8	0.524	6
Social	11.805	5	0.524	6
Government	9.5752	6	0.524	6
Education	4.4658	9	0.524	6
Environment	1.9131	14	0.524	6
Health	0.5849	21	0.524	6
Economy	17.793	4	0.4469	13
Transport	4.4252	10	0.4469	13
Architecture	3.0277	13	0.4469	13
Media and art	0.696	19	0.3569	16
Standardization	1.1053	16	0.3402	17
Safety	0.9881	17	0.3273	18
Energy	1.5842	15	0.3273	18
Tourism	0.3064	22	0.2502	20
Waste	0.8745	18	0.2502	20
History	0.6655	20	0.1974	22

Note: Abbreviations: WD = Weighted Degree, and EC = Eigenvector Centrality. The shades illustrate the rising elements in eigenvector centrality compared with weighted degrees.

. The shades in Table 5 illustrate the rising elements in eigenvector centrality compared with weighted degrees. These elements include private, people, academic, and NGO sectors, and the services concerning data, education, environment, health, media and art, and tourism. The intermediated elements are shown to increase their connectivity and importance in connected sustainable smart cities. The connected, sustainable smart cities emphasize the human resources or agencies to connect services through their data and active participation. Private sectors play a crucial role in attracting innovations by collaborating with other stakeholders to provide various services. They cooperate with academia to research new technologies and services based on data provided by citizens and other organizations [63]. Smart cities have been developed in cooperation with the public and private sectors. From the perspectives of governance systems, local government partnerships, and supportive national governments are necessary to build on private creative ideas from private sectors and people who own successive localities from generation to generation. The local government subsidies contribute to the advancement of local communities and private sectors. Devolution strengthens and amplifies the networks among infrastructure, data, social governance, education and environment, and health, which are the result of this research. Social services interconnect private sectors with other sectors. The other four services, including data, government, environment, and health services, provide fundamental linkages between the public and private sectors. In this sense, city networks, which are built up on a foundation of developing local entities, transform the service ecosystems to bring out intermediated services, even though the cities are not geographically adjacent.

5. Discussion

The first result reflects the urban geographic economy and demonstrates the distinct features of conceptually related smart cities as they undergo socio-technical transitions through measures such as weighted degree, eigenvector centrality, and betweenness centrality. The context of smart city implementation plays a significant role in current developments, with high-weighted services reflecting the earliest implemented services, while the latter services are essential for connecting with existing or emerging services mediated by stakeholders. Data and knowledge are among the intermediated services that have recently gained importance amid socio-technical transitions. The concept of eigenvector centrality, betweenness centrality, and weighted degrees emerged in urban geography. The eigenvector centrality is correlated with self-reinforcement in the urbanization economy, while betweenness centrality is the intermediate element. As the urbanization economy shares some intermediate elements, such as business services, transportation services, public infrastructure, and labor pooling, organizations that require face-to-face contact, including corporate headquarters or knowledge-based businesses, tend to cluster as self-reinforcing factors [116]. This means that the intermediate elements are crucial to the modern economy [116]. In this paper, the services with weighted degrees include fundamental services such as infrastructure, economy, social, data and government, while the service with high betweenness centrality is data, which connects the high-weighted services so that it has high eigenvector centrality. In this light, services with high betweenness centrality have the potential to drive emerging industries or services, similar to how it occurs in the urban geographic economy. The use of connected intelligent data, including artificial intelligence, can serve as a unifying force to link urban services, living organisms, organizations, and environments within a governance model in order to create a safer and more prosperous world in densely populated and centralized areas [59,117]. Accordingly, the ecosystems of evolving service systems are identified through the geographic economy. When it extends to the relationship among geography, organization and specific fields, Ma (2023) clarifies multi-proximity factors driving dynamics, including geographical proximity, research contextualized cognitive proximity, and organizational proximity [105]. The first result extends the existing literature by clarifying the initial implemented services frame influence on the later network of services. It highlights the necessity of understanding the context of smart city services evolution in the context of socio-technical transitions for making sustainable smart cities. Moreover, the study indicates that the smart cities’ service development ecosystems are analogous to the urban geographic economy regarding the relationships among stakeholders or organizations, intermediated services, and self-reinforced services in the urbanization economy.

The second finding underscores the importance of multi-stakeholder partnerships from perspectives of service development. It presents various stages of developing conceptually related smart city services, depending on stakeholder partnerships. Public–private partnerships have been increasingly utilized in the implementation of smart city services and urban planning in recent years. The concept of public–private partnerships in urban governance, which is influenced by neo-liberalization, intends to achieve a common goal, often in the form of infrastructure development or service provision. While the public private partnerships have the potential to bring innovation and efficiency to urban planning, there are also concerns that they may prioritize profit over public interest and may not adequately address issues of social equity and environmental sustainability [118]. In this context, the concept of communicative planning has emerged as an alternative approach that seeks to involve diverse stakeholders in decision-making processes. This approach prioritizes inclusivity and seeks to ensure that all voices are heard and taken into consideration in the planning process. By doing so, communicative planning can help to avoid the isolation of vulnerable ecosystems and species and promote more sustainable and equitable outcomes. Moreover, multi-stakeholder partnerships have the potential to offer an alternative solution to emerging challenges of urbanization, particularly regarding the issues of energy and waste. The telecommunication fields endeavors to establish dual systems of smart city service systems between top-down and bottom-up that are ontology-based systems mediated by data to provide improved services to all stakeholders with limited resources [63]. Urban planning balanced between top-down and bottom-up approaches has the potential to provide solutions to the future smart city challenges by encouraging citizens to interconnect with urban systems and organizations through mobilizing sustainable smart cities based on vision and measurable and controllable elements in master plans [119]. The diverse stakeholder partnerships include cooperation among public, private, academic, and NGOs sectors and embrace the concept of devolution or decentralization. Devolution, referring to the definite ownership and self-responsibility of investment from a city or a region’s ownership, contributes to increasing data awareness among stakeholders [59]. The connectivity and growing power of regions or cities lead to devolution, as occurred in the 20th century in response to colonization [120]. The sustainable development goals indicate two innovation approaches which are gaining prominence, including the vital role of local leaders to drive global change in the transformative power of urbanization (equity innovation with multi-stakeholders input), and in leaving no one behind (inclusive innovation) [4]. The interactions with local government foster connectivity with more local players, leading to prompt and suitable actions toward local challenges with voluntarily participation from diverse stakeholders that results in improvements in the services and connectivity with decision making processes [121]. The New Urban Agenda points out the appropriately balanced governance systems among the national government, subnational and local governments, and relevant stakeholders to revitalize, strengthen and create partnerships [122]. In this sense, this study provides a managerial contribution regarding what types of partnerships are appropriated for European sustainable smart cities to promote specific services. Furthermore, it empirically demonstrates the necessity of multi-stakeholder partnerships for making sustainable smart cities.

The last finding concerns the city network. The concept of a city network has been discussed in line with glocalization, a concept that appeared in the Harvard Business Review [123], conurbation as mentioned by Patrick Geddes [124], and city knowledge exchanges in city expositions or exhibitions. City networks have been traditionally considered among geographically adjacent cities, such as in the concepts of conurbation and decentralization underlying the concept of city-regional development. However, with emerging smart city development, the direction has changed into networking for improving governance systems and technologies, taking into account existing assets, budgets, challenges and the background of socio-technical urban transitions. A typical instance is a city memorandum of understanding between public organizations or public and private organizations. Calzada (2017) argues that cities have the power to compete or cooperate as an investment destination so that the national government is not necessarily distributing subsidies for them [59]. Notably, the private sector has been at the forefront of digital transformation, especially during the COVID-19 era [4]. This is based on local governmental supportive investment and legislation. The United Nations highlights that the next generation of digitalization requires an ecosystem-centric approach in which the public sector plays an entrepreneurial role in spurring innovation with private sectors based on fruitful research in high-growth and high-risk areas and bringing diverse stakeholders for long-run growth strategies [4]. Serrano et al. (2020) raise the issue that smart city networks can become a regional gateway to expand the business of multinational firms rather than empowering medium-sized cities or small national firms. In this regard, the local government should take an entrepreneurial role to empower local communities and corporations to germinate local innovation, and expand their influences globally with other smart cities as a form of multinational organization. This paper empirically identifies the emerging stances of private sectors under the assumption of connected cities, services, and stakeholders as a whole in sociotechnical transitions.

6. Conclusions

Sustainable Smart cities are multi-faceted living ecosystems developed through diverse stakeholder participation in sociotechnical transitions. However, the lack of a suitable governance model that incorporates the various components of these systems, such as connected technologies, data, services, stakeholders, organizations, and legislation, has led to indiscriminate development with a variety of names and oversimplification of technologies without consideration of the local context and traits of smart urbanization in sociotechnical transitions. This study aims, from the governance and sociotechnical systems perspectives, to identify the characteristics of conceptually related smart city services implementations depending on stakeholder partnerships. To achieve this goal, this study has narrowed down three objectives: (1) to clarify the characteristics of services in the evolution of conceptually related smart cities by expanding on the existing literature, (2) to demonstrate the different phases of developing conceptually related smart cities services depending on different stakeholder partnerships, and (3) to identify connected services and stakeholders, assuming that conceptually related smart cities are connected virtually and physically.

The application of social network analysis illustrates the relationships among stakeholders, services, and cities in establishing a smart governance model. The data for the method is based on the findings of Kim and Yang’s (2023) study [34], as their objectives and ideas pertaining to sustainable smart cities align with this study. The target cities selected for analysis are European sustainable smart cities, given Europe’s continued leadership in e-government development, as evidenced by its consistent top ranking in the United Nations e-government development index (EGDI) since 2010. Specifically, Barcelona, Berlin, and London were chosen as they exemplify the European cases for the operational definition of sustainable smart cities used in this study. The dataset on stakeholders, services, and cities reveals several key findings. Firstly, the initial services associated with the conceptually related smart cities are reflected in the accumulated and current characteristics of the smart city services, depending on stakeholder partnerships. However, the network features are different between the initial and later services. Secondly, the development of different services depends on stakeholder partnerships, indicating that multiple stakeholders, including local entities, must establish partnerships to tackle the current challenges of massive urbanization. Finally, the analysis highlights the growing presence of private sectors and intermediate services in the global network of cities.

This study is subject to certain limitations despite the sophisticated structures utilized to demonstrate the empirical CRSCs service evolutionary characteristics. Firstly, there is little explanation provided for how services are adapted to different geographical contexts based on stakeholder partnerships by maximizing societal benefits without incurring negative externalities. This could be addressed by consulting the existing literature, particularly Kim and Yang’s (2023) study [34]. Secondly, the study lacks a temporal or geographical dimension, which could be remedied by researching geographically networked services among organizations with yearly or monthly data utilizing geographic information systems. Thirdly, the results may not be readily generalizable, as they are based on data from only three cities. To address this issue, it is recommended to analyze an identical number of sample cities in each continent, using identical methodologies for data collection, sorting, coding, classifying, and analysis, with periodic matrix taxonomy and social network analysis. Lastly, this study does not serve to meet the urgent requirement in respect of implementation of the governance model or the provision of pragmatic specifications for the most effective ICT investments to be made. Nor does it contribute to the decision-making process addressing the grand challenges for Europe and global cities in meeting political commitments regarding climate change mitigation and adaptation. Further studies need to address these issues by conducting qualitative in-depth research to investigate the significant challenges confronting in European and global megacities.

Nonetheless, the results of this study have significant managerial implications, as they enable the identification of elements with high eigenvector, intermediate elements, and highly implemented fundamental services, depending on different stakeholder partnerships. These findings can inform decision-making regarding services development and contribute to the development of new smart cities by creating a smart city governance model with multi-faceted, multidisciplinary, and multilevel systems of stakeholder sectors and services, all connected by multiple partnerships. Additionally, this study has theoretical implications, as it empirically demonstrates the necessity of multi-stakeholder partnerships and devolution to build sustainable smart cities. Overall, this research can help to advance the understanding of smart city development, contributing to the practical and theoretical discourse on the subject.