Developing a Composite Sustainable Smart City Performance Assessment Index: A Novel Indexing Model and Cross-Country Application

Abstract

Cities are increasingly expected to address digital transformation and sustainability challenges at the same time. However, existing urban indices generally approach smart city and sustainable city perspectives separately, which limits their ability to capture the integrated nature of contemporary urban development. In addition, many index-based studies rely on similar methodological choices. This study develops a composite Sustainable Smart City (SSC) index supported by a systematic scoring framework that brings smartness and sustainability together. The proposed framework follows a step-by-step procedure covering data preparation, normalization, weighting, aggregation, and final scoring. To address information overlap among indicators, a Redundancy-Penalized Entropy Weighting (RPEW) approach is applied. Then, overall SSC scores are calculated using a soft non-compensatory aggregation to emphasize balanced performance across dimensions. The framework is empirically illustrated through a cross-country case study including 38 OECD (Organization for Economic Co-Operation and Development) countries. A machine-learning-based polynomial forecasting approach is used for a limited number of indicators to deal with data gaps allowing the assessment to reflect more up-to-date conditions. The results highlight clear differences in SSC performance and show that strong outcomes in a single dimension are not sufficient to achieve high overall SSC scores. Instead, balanced progress across economic, digital, environmental, governance, mobility, and social dimensions plays an important role. In addition, the proposed framework provides a practical basis for comparative analysis, benchmarking, and policy-oriented evaluation of smart and sustainable urban development.

1. Introduction

Cities are facing increasing pressure due to rapid urbanization and growing resource demands. Urban populations continue to grow. This population growth creates significant problems for infrastructure, environmental protection, and public service delivery. At the same time, cities must respond to climate change, energy transition, and social inequality to improve citizens’ quality of life [1,2]. These issues make urban management more complex than ever before. In recent years, digital technologies have also become an important part of urban development. Cities are investing in data systems, digital infrastructure, and new technological solutions. These tools support transportation management, energy efficiency, environmental management and public administration [3,4]. As a result, urban development is increasingly influenced by both technological progress and sustainability concerns. Because of these changes, cities can no longer be evaluated using single indicators or narrow performance measures. Urban systems involve economic, environmental, social, and technological dimensions that interact with each other. For this reason, researchers and policymakers increasingly rely on multidimensional indicator frameworks to understand urban performance and guide policy decisions [5].

The smart city concept has emerged as a response to the growing complexity of urban systems. This concept generally refers to cities that use digital technologies and data-driven solutions to improve urban management, quality of life and service delivery. Information and communication technologies allow cities to efficiently manage urban processes and infrastructure. These technologies also support new forms of governance and decision-making [6,7]. However, smartness is not limited to technological infrastructure. Many studies emphasize the role of innovation capacity, knowledge creation, and digital connectivity in shaping smart urban development. Cities that are able to generate, adopt, and diffuse knowledge are better positioned to respond to rapidly changing urban conditions. In this sense, smart cities are often described as environments where technology, innovation, and human capital interact to improve urban performance and quality of life [8,9]. In this study, smartness is defined as the ability of a city to create, adopt, and utilize knowledge, innovation, and digital technologies to improve urban management and enhance the delivery of public services.

Sustainability also plays a central role in contemporary urban development. The main goal of sustainable cities is to ensure long-term urban resilience while protecting the environment and improving social well-being in an economic manner. Considering these points, urban sustainability requires efficient resource use, reduced environmental impacts, and inclusive social development. Cities must manage energy consumption, environmental quality, and social equity while maintaining economic vitality [10,11]. In this study, sustainability is defined as the ability of a city to maintain environmental quality, support economic prosperity, and promote social well-being while ensuring the responsible use of natural resources and long-term urban resilience.

Smart city initiatives mainly focus on technological innovation and digital infrastructure. But this focus is not enough for sustainable urban development. Smart technologies can improve the efficiency of urban systems. Yet, they do not solve environmental and social problems. For this reason, many studies emphasize that smart city policies should be combined with sustainability objectives. The Sustainable Smart City (SSC) concept has emerged from these two perspectives, smartness and sustainability. SSC approaches aim to integrate digital technologies with environmental protection, social well-being, and economic development. In this way, technological innovation can contribute not only to urban efficiency but also to long-term sustainable urban development [12,13,14].

In response to the growing complexity of urban systems, several index-based approaches have been developed to evaluate city performance. Many of these indices focus either on smart city development or on urban sustainability. Smart city indices generally emphasize technological infrastructure, digital connectivity, and innovation capacity [5,15]. The literature also includes several studies that compare different smart city assessment frameworks and indicator sets in order to evaluate their scope, methodology, and practical applicability [16,17]. Sustainable city indices typically focus on environmental performance, resource efficiency, and social well-being [18,19]. In recent years, some studies have attempted to combine these two perspectives under the concept of Sustainable Smart City (SSC). However, the number of index studies that explicitly integrate both smart and sustainability dimensions remains relatively limited [20,21,22]. Most existing indices also follow a similar methodological structure that includes indicator selection, normalization, weighting, and aggregation. In many cases, the selection of indicators is not clearly justified and indicators are assigned equal weights regardless of their informational contribution. Conventional normalization and aggregation approaches are frequently applied without considering the influence of outliers or scale differences across indicators. In addition, potential redundancy among indicators is often ignored. These issues may reduce the transparency and interpretability of index results and limit the ability of existing indices to fully capture the multidimensional nature of sustainable smart city development.

The motivation of this study is to address these gaps by proposing a systematic framework for scoring Sustainable Smart Cities (SSCs). Rather than treating smartness and sustainability as independent constructs, this research brings them together within an index structure. The proposed framework is designed to ensure methodological coherence across all stages of index construction, from data preparation to final scoring.

Based on this motivation, the study is guided by the following research questions:

RQ1:

How can smart and sustainable city dimensions be systematically integrated into a single composite index without losing their individual analytical meaning?

RQ2:

To what extent does a structured and transparent scoring framework improve the comparability of SSC performance across countries?

To answer these questions, this study makes several contributions to the literature. First, it develops an SSC index that brings smart and sustainable dimensions together using 55 indicators under 7 dimensions. Second, it proposes a systematical methodological framework for SSC scoring. Third, the framework is applied to a cross-country case study, demonstrating its practical usability and analytical value. Finally, this study also uses a forecasting approach to estimate unavailable values for the most recent year. By integrating forecasted data into the index calculation, the proposed framework allows for up-to-date SSC performance assessment while maintaining methodological consistency. In addition, robustness analysis is conducted to evaluate the sensitivity of the results to alternative weighting approaches.

This study is organized as follows. Section 2 provides an extensive literature review on indices. Section 3 represents the proposed methodology. The application of the proposed SSC Indexing Model is illustrated in Section 4. Section 5 discusses the results and explains the managerial implications. Lastly, conclusions are given in the last section.

2. Literature Review

In recent years, city indices have been widely used to evaluate urban performance in the contexts of digital, smart and sustainable cities. These indices aim to combine different aspects of urban systems into a single structure that allows comparison across cities or countries. As cities face increasing pressure from rapid urbanization, digital transformation, and environmental challenges, such index-based approaches have become more common in both academic studies and policy reports.

In the literature, studies can be grouped into two main categories. The first focuses on smart city indices, which mainly assess technological capacity, digital infrastructure, innovation, and data-driven urban services [5,15]. These indices usually emphasize efficiency, connectivity, and technological advancement. The second group consists of sustainable city indices, focusing on environmental protection, social well-being, resource efficiency, and long-term resilience [18,19]. These approaches are often closely linked to sustainability agendas and development goals [23].

While smart city and sustainable city indices provide useful insights, they are generally developed separately. As a result, they tend to capture only part of the broader urban development picture. Smart city indices often give limited attention to environmental and social issues, whereas sustainable city indices frequently overlook digitalization, technology and innovation aspects. This separation makes it difficult to evaluate cities in a multi-dimensional way.

The concept of the SSC has emerged as an attempt to address this limitation by bringing smartness and sustainability together within a single framework. However, indices that explicitly adopt an SSC perspective are still very limited in number. Moreover, existing SSC-oriented indices differ significantly in their structure and scope, and many of them do not fully capture the complexity of SSC development. In several cases, the balance between smart and sustainable dimensions remains weak, or the methodological structure is not sufficient or is not provided for assessments.

Given this situation, reviewing existing smart city, sustainable city, and SSC-related indices is an important step for understanding current studies and identifying their limitations. This review makes it possible to compare indicator selection procedures and methodological frameworks.

To identify relevant studies and index frameworks, Web of Science database was used. The search included combinations of keywords such as “smart city index”, “sustainable city index”, “smart sustainable city index”, “sustainable smart city index”, and “urban composite index”. To review widely recognized institutional index reports websites and publicly available sources on the internet were also used. This step was necessary because several widely used city indices are published as institutional reports rather than peer-reviewed academic articles. Based on this process, representative indices were selected for further analysis according to their relevance to smart city or urban sustainability assessment, transparency of methodology, and availability of indicator structures.

The following tables present well-known indices from the literature and they summarize their main characteristics. Dimensions and indicators of IESE Cities in Motion Index [24] are shown in Table 1.

Table 1. Dimensions and Indicators of IESE Cities in Motion Index [24].

Dimensions	Indicators
Human Capital	Secondary and higher education, Schools, Business schools, Expenditure on education, among others.
Social Cohesion	Female-friendly, Hospitals, Crime rate, Slavery Index, Happiness Index, Gini Index, and related social well-being indicators.
Economy	Unicorn companies, Ease of starting a business, Global Startup Ecosystem Index, Mortgage, among others.
Governance	Bitcoin legal, ISO 37120 certification [25], Government buildings, Embassies, and related institutional indicators.
Environment	CO2 emissions, Methane emissions, Environmental Performance Index, Pollution Index, among others
Mobility and Transportation	Bicycle, moped or scooter rental service, Bike sharing, Metro stations, Traffic Inefficiency Index, and related accessibility indicators.
Urban Planning	Bicycles, Bike Advance, Buildings, among others.
International Profile	Number of passengers per airport, Hotels, Restaurant Price Index, McDonald’s, Number of congresses and meetings.
Technology	Mobile broadband, Innovation Cities Index (ICI), Internet, Computers/PCs, among others.

Note: Each dimension of the IESE Cities in Motion Index [24] includes several sub-indicators. Only representative examples are shown for brevity.

The IESE Cities in Motion Index [24] evaluates 183 global cities across nine dimensions, including economy, human capital, international profile, urban planning, environment, technology, governance, social cohesion, and mobility and transportation. Each indicator’s relative weight is determined using the complement of its coefficient of determination (R²) with respect to the rest of the indicators. It ensures that highly correlated variables have a smaller impact.

Dimensions and indicators of Arcadis Sustainable City Index [26] are shown in Table 2.

Table 2. Dimensions and Indicators of Arcadis Sustainable City Index [26].

Dimensions	Indicators
Planet	Air pollution, Drinking water and sanitation, Energy, Natural disaster resilience, Green spaces, Greenhouse gas emissions, Green policy, Sustainable mobility, Waste management.
People	Public transport services, Affordability, Crime, Cultural offerings, Education, Health, Income inequality, Work–life balance.
Profit	Access to reliable electricity, Quality of internet, Ease of doing business, Economic development, Income and living standards, Employment, Governance and digital services, Transport infrastructure.
Progress	Public transport services, Education, Health, Income inequality, Air pollution, Drinking water and sanitation, Energy, Income and living standard, Employment.

The Arcadis Sustainable Cities Index [26] aims to evaluate 100 global cities under 4 dimensions to provide a benchmarking framework for policy makers and urban planners. Each pillar is assigned an equal weight and normalized values are achieved using min-max method. The overall city score is calculated as the arithmetic mean of the four equally weighted pillars.

Table 3 represents indicators of the Corporate Knights Sustainable Cities Index [27].

Table 3. Indicators of the Corporate Knights Sustainable Cities Index [27].

Indicators
Scope 1 GHG emissions
Consumption-based emissions
Particulate air pollution
Open public space
Water access
Water consumption
Automobile dependence
Road infrastructure efficiency
Sustainable transport
Solid waste generated
Climate change resilience
Sustainable policies

The Corporate Knights Sustainable Cities Index [27] evaluates 70 global cities through 12 quantitative indicators. Each indicator is normalized on a 0–1 scale relative to the best-performing city. A distinctive feature of the index is the Corporate Knights Socio-Economic Adjustment Factor (CKSEAF), which adjusts environmental scores according to the socio-economic context of each city, using human development index (HDI), income equality (1–GINI), and GDP per capita as adjustment parameters. The overall city score is calculated as the weighted sum of these adjusted indicators, with weights determined through expert evaluation.

Dimensions and indicators of Sustainable Development Report Index [28] are shown in Table 4.

Table 4. Dimensions and Indicators of Sustainable Development Report Index [28].

Dimensions	Indicators
No Poverty	Poverty headcount ratio at $2.15/day, Poverty headcount ratio at $3.65/day, Poverty rate after taxes and transfers.
Zero Hunger	Prevalence of undernourishment, Prevalence of stunting in children under 5 years of age, Prevalence of wasting in children under 5 years of age, and related nutrition indicators.
Good Wealth and Well-Being	Maternal mortality ratio, Incidence of tuberculosis, Traffic deaths, among others.
Quality Education	Participation rate in pre-primary organized learning, Literacy rate, Net primary enrollment rate, and similar educational indicators.
Gender Equality	Ratio of female-to-male mean years of education received, Seats held by women in national parliament, Demand for family planning satisfied by modern methods, and related gender measures.
Clean Water and Sanitation	Population using at least basic sanitation services, Scarce water consumption embodied in imports, Population using safely managed sanitation services, among others.
Affordable and Clean Energy	Population with access to electricity, Population with access to clean fuels and technology for cooking, and associated energy indicators
Decent Work and Economic Growth	Adjusted GDP growth, Victims of modern slavery, Employment-to-population ratio, and related economic indicators.
Industry, Innovation and Infrastructure	Population using the internet, Mobile broadband subscriptions, Articles published in academic journals, among others.

Note: Each dimension of the Sustainable Development Report Index [28] includes multiple sub-indicators aligned with the 17 Sustainable Development Goals (SDGs). Only representative examples are shown for brevity.

The Sustainable Development Report assesses 167 countries using 125 indicators aligned with the 17 SDGs. Indicator values are normalized through a min–max transformation to express performance as a distance to target on a 0–100 scale. Goal-level scores are computed as the arithmetic mean of all indicators within each SDG. The overall SDG Index score for each country is obtained as the arithmetic mean of the 17 goal scores. Equal weighting is applied across all goals [28].

Another index developed by the United Nations is the City Prosperity Index (CPI) [29], introduced by UN-Habitat. The CPI is designed to enable city authorities, as well as local and national stakeholders, to identify opportunities and potential areas of intervention to enhance urban prosperity. The composite index is structured around six dimensions: productivity, infrastructure development, quality of life, equity and social inclusion, environmental sustainability, and governance and legislation. These dimensions provide a comprehensive framework for defining targets and goals that support the formulation of evidence-based policies, including the development of city visions and long-term strategies that are both ambitious and measurable. By 2020, the CPI had been applied to evaluate urban performance in 539 cities across 54 countries worldwide. The dimensions and indicators of the CPI [29] are provided in Table 5.

Table 5. Dimensions and Indicators of the CPI [29].

Dimensions	Indicators
Productivity	Economic growth, Employment, Economic agglomeration
Infrastructure and Development	Housing infrastructure, ICT, Urban mobility
Quality of Life	Health, Education, Safety and security, Public space
Equity and Social Inclusion	Economic equity, Social inclusion, Gender equity
Environmental Sustainability	Air quality, Waste management, Energy/Climate change
Urban Governance and Legislation	Participation and accountability, Municipal finance, Legal framework

The dimensions and indicators of the IMD Smart City Index [30], a smart city assessment framework, are presented in Table 6.

Table 6. Dimensions and Indicators of the IMD Smart City Index [30].

Dimensions	Structures	Technologies
Health & Safety	Basic sanitation meets the needs of the poorest areas, Recycling services are satisfactory, Public safety is not a problem, among others.	Online reporting of city maintenance problems provides a speedy solution, A website or App allows residents to easily give away unwanted items, Free public Wi-Fi has improved access to city services, among others.
Mobility	Traffic congestion is not a problem, Public transport is satisfactory.	Car-sharing Apps have reduced congestion, Bicycle hiring has reduced congestion, and related accessibility indicators.
Activities	Green spaces are satisfactory, Cultural activities (shows, bars, and museums) are satisfactory.	Online purchasing of tickets to shows and museums has made it easier to attend.
Opportunities (Work & School)	Employment finding services are readily available, Most children have access to a good school, Lifelong learning opportunities are provided by local institutions, among others.	Online access to job listings has made it easier to find work, Online services provided by the city has made it easier to start a new business, among others.
Governance	Information on local government decisions are easily accessible, Corruption of city officials is not an issue of concern, and related institutional indicators.	Online public access to city finances has reduced corruption, Online voting has increased participation, among others.

Note: Each dimension of the IMD Smart City Index [30] includes several sub-indicators. Only representative examples are shown for brevity.

The IMD Smart City Index [30] is used widely by academics [31]. It aims to measure how effectively a city uses technology to improve its citizens’ quality of life. It applies a combined weighting and scoring approach based on survey data collected from 120 residents per city. Results from the 2024, 2023, and 2021 editions are merged using a 3:2:1 temporal weighting ratio to account for data stability and recent developments. Within each of the two main pillars—Structures and Technologies—indicators are equally weighted across five thematic areas (health and safety, mobility, activities, opportunities, and governance). Scores are normalized to a 0–100 scale within each HDI group, and the final index score for each city is calculated as the arithmetic mean of the two pillar scores.

Indicators and dimensions of Global Power City Index [32] are demonstrated in Table 7.

Table 7. Dimensions and Indicators of the Global Power City Index [32].

Dimensions	Indicators
Economy	Nominal GDP, GDP per capita, Economic freedom, and related economic indicators.
Research and Development	Number of researchers, World’s top universities, Research and Development (R&D) expenditure, among others.
Cultural Interaction	Number of international conferences, Number of cultural events, Cultural content export value, and similar cultural engagement indicators.
Livability	Total unemployment rate, Total working hours per capita, Workstyle flexibility, Housing rent, among others.
Environment	Commitment to climate action, Renewable energy rate, Waste recycle rate, and related environmental indicators.
Accessibility	Cities with direct international flights, International freight flows, Number of air passengers, and associated transport indicators.
Economy	Nominal GDP, GDP per capita, Economic freedom, and related economic indicators.

Note: Each dimension of the Global Power City Index [32] includes several sub-indicators. Only representative examples are shown for brevity.

The Global Power City Index 2025 [32] measures 48 cities through 70 quantitative and qualitative indicators grouped under six dimensions: Economy, research and development, culture, livability, environment, and accessibility.

Table 8 presents India Smart City Index [33] structure.

Table 8. Dimensions and Indicators of India Smart City Index [33].

Dimensions	Indicators
Living	Access to electricity, Access to water supply, Maternal health, Crime incidence, among others.
Economy	Unemployment rate, GDP, Income distribution, Growth of new businesses, Workforce participation of women.
Governance	Water distribution efficiency, Planning framework, Ease of access to government services, among others.
People	Higher education, Gender inclusivity, Engagement with city administration, among others.
Mobility	Share of green modes, Road Safety, Trip length, and related accessibility indicators.
Environment	PM2.5 concentration, Use of renewable energy, Solid waste recycling, Sewage recycling.
Living	Access to electricity, Access to water supply, Maternal health, Crime incidence, among others.

Note: Each dimension of the India Smart City Index [33] includes several sub-indicators. Only representative examples are shown for brevity.

The India Smart Cities Index [33] evaluates 53 Indian cities across six key dimensions: Living, Governance, People, Economy, Mobility, and Environment. Each dimension consists of multiple sub-factors represented by 58 measurable indicators. The index applies a decile-based scoring approach and it converts indicator values into ten-point scale scores (1–10) according to their percentile ranks across cities. Equal weighting is applied to all indicators, and each dimension score is obtained as the arithmetic mean of its corresponding indicator scores. The overall city score is then calculated as the average of the six-dimension scores, ensuring a balanced and comparable evaluation of smart city performance within the Indian urban context.

Tas and Alptekin [34] developed a Smart City Index for 30 metropolitan municipalities in Turkiye using a hybrid multi-criteria decision-making framework. Their indices’ indicators are shown in Table 9.

Table 9. Dimensions and Indicators of Tas and Alptekin’s Index [34].

Dimensions	Indicators
Livability	Number of population, Life expectancy, Population density, Life satisfaction “Very happy” rate
Economy	Number of venture, Employment rate, Per capita GDP
Environmental Sustainability	Average municipal waste per capita, Total electricity consumption per person, Ratio of municipality population served by wastewater treatment plant to total municipality population, Forest area ratio
Research and Development	Number of university students, Number of fiber internet subscribers, Fiber-Optic cable length
Accessibility	Number of motor vehicles, Number of traffic accidents
Cultural Interaction	Number of official tourism facilities, Number of visitors to museums and ruins affiliated to the ministry, Number of bed place 2020

The study applied the MEREC (Method Based on the Removal Effects of Criteria) method to determine objective indicator weights and the MARCOS (Measurement of Alternatives and Ranking According to Compromise Solution) method to rank cities based on overall smart city performance. The index structure consists of six main dimensions—Livability, Economy, Environmental Sustainability, Research and Development, Accessibility, and Cultural Interaction—consisting of 19 indicators [34].

Pira [20] introduces an integrated SSC framework that brings together the core ideas of smart city initiatives and sustainability. Dimensions and indicators of Milad Pira’s SSCs Index are represented in Table 10.

Table 10. Dimensions and Indicators of Milad Pira’s SSCs Index [20].

Dimensions	Indicators
Socio-cultural	Healthcare delivery, Quality drinking water, individuals’ health monitoring, Quality food, Education funding, Free education, Low crime rate, Population density, Population growth rate, investment in culture, Civic engagement
Economic	Affordable housing, Start-ups, International collaboration, Low poverty rate, Job opportunities
Environmental	Green spaces, Air quality, Low pollution, Energy use, Waste generation, Sustainability-certified buildings
Governance	E-governance, Real-time data monitoring, Internet and Wi-Fi coverage, Disaster preparedness, Public transport, Clean-energy transport

Using a two-stage content analysis, the study examines where the Smart City Index master indicators and the OECD (Organization for Economic Co-Operation and Development) Sustainable Development Indicators intersect and align. Based on this analysis, a unified set of 28 indicators is proposed, grouped into four broad dimensions: socio-cultural, economic, environmental, and governance [20].

Another study has been conducted by Gazzeh [35] to prioritize a core set of SSC indicators using a combined methodology: content analysis and Analytic Hierarchy Process. Dimensions and indicators of Gazzeh’s SSC Index are represented in Table 11.

Table 11. Dimensions and Indicators of Gazzeh’s [35] SSC Index.

Dimensions	Indicators
Living	Quality of life, Healthcare services & conditions, Housing quality & Harmonious Living, among others.
People & Society	Social and Human Capital Qualification Level, Education System and Facilities Cosmopolitanism, Social & cultural plurality and inclusion, among others.
Environment	Sustainable resource management, Attractiveness of Land & green environment, and Public spaces, Environmental quality & protection, among others.
Infrastructure	Access to improved water, Drainage, Access to improved sanitation, Solid waste treatment
Economy & Productivity	Competitiveness, flexibility, economic vitality, Innovation & Knowledge-based economy, E-commerce, among others.
Governance	Participation in decision-making, Accountability, Justice & Fairness, among others.
Mobility & Transportation	Travel distance, Public transport alternatives, Accessibility, among others.
Technology & ICT	Use of ICT in infrastructure management, Quality of municipality websites, Business grade wifi hotspots, among others.

Note: Each dimension of Gazzeh’s [35] SSC Index includes several sub-indicators. Only representative examples are shown for brevity.

Dimensions and indicators of Ozkaya and Erdin’s [21] SSC Index are represented in Table 12.

Table 12. Dimensions and Indicators of Ozkaya and Erdin’s [21] SSC Index.

Dimensions	Indicators
Smart economy	Innovation, R&D, Entrepreneurship, Headquarters, among others.
Smart people	Universities, Participate in education, English Proficiency, Lifelong learning, among others.
Smart governance	Electoral process and pluralism, Political participation, Civil liberties, Female city representatives, Government effectiveness, among others.
Smart mobility	Local accessibility, Freedom and openness, ICT access, Traffic safety, among others.
Smart environment	Sunshine hours, Public green space share, Environmental protection, Efficient use of water, among others.
Smart living	Cultural interaction, Health, Affordability, International inbound tourist, among others.

Note: Each dimension of Ozkaya and Erdin’s [21] SSC Index includes several sub-indicators. Only representative examples are shown for brevity.

Ozkaya and Erdin [21] evaluated smart and sustainable cities under 6 dimensions and 47 indicators. Their study uses Analytical Network Process (ANP) to weigh smart and sustainable city criteria. The study covers 44 cities around the world and comparisons were made by TOPSIS (Technique for Order Preference by Similarity to Ideal Solution).

Dimensions and indicators of Shmelev and Shmeleva’s [22] SSC Index are represented in Table 13.

Table 13. Dimensions and Indicators of Shmelev and Shmeleva’s [22] Smart and Sustainable City Index.

Dimensions	Indicators
Economic	Gross regional product, Income, Inflation, Unemployment
Smart	Patents, Internet, Metro
Social	Life, Education, Gini index
Environmental	CO2, Renewables, PM110, Water, Waste, Recyling

To assess urban sustainability performance, this study applies a multi-criteria approach using a panel of 20 indicators to a set of 57 global cities. Various indicator weightings produced aggregate performance scores for global cities under four dimensions: economic, social, environmental and smart [22].

Haksevenler et al. [18] proposed a sustainable city index using United Nation’s [36] 17 Sustainable Development Goals (SDGs). Dimensions and indicators used in their study are shown in Table 14.

Table 14. Dimensions and Indicators of the Sustainable City Index proposed by Haksevenler et al. [18].

Dimensions	Indicators
Environmental	Net carbon footprint, Air quality index, Ecological status, Exposure to disasters
Social	Average traffic rate in rush hours, Education level, Rate of chronic disease, Women’s social status, Quality of life,
Economic	Resource consumption, Socio-economic level, Rate of receiving social assistance, Rate of car ownership, Income level, Accessibility to communication networks
Institutional	Management scorecard value, Satisfaction with institutions, Amount of open green space, Waste recycling rate, Access to public transportation, Access to healthcare institutions

This study [18] integrates SDGs as indicators and creates a sustainable city index consisting of 4 dimensions: environmental, social, economic, and institutional. Min-max normalization procedure, equal weighting and weighted sum aggregation methods have been selected to assess districts of Istanbul. Then, various multi-criteria decision-making methods have been used to analyze robustness of the methodology.

Cohen [15] defines smart city concept under 6 dimensions and 18 working areas. Dimension and indicators used in Cohen’s smart city index are represented in Table 15.

Table 15. Dimensions and Working Areas of the Smart City Index proposed by Cohen [15].

Dimensions	Working Areas
Smart economy	Entrepreneurship & Innovation, Productivity, Local and global connection
Smart people	Inclusion, Education, Creativity
Smart governance	Online services, Infrastructure, Open Government,
Smart mobility	Efficient transport, Multi-model access, Technology infrastructure,
Smart environment	Smart buildings, Resources Management, Sustainable Urban Planning
Smart living	Culture and well-being, Safety, Health

Another smart city index has been introduced by Abu-Rayash and Dincer [5]. The index consists of eight dimensions and 32 indicators. Using four distinct weighing schemes, the model is applied to 20 cities worldwide, ranked for their smart performance. Their smart city index is shown in Table 16.

Table 16. Dimensions and Indicators of the Smart City Index proposed by Abu-Rayash and Dincer [5].

Dimensions	Indicators
Smart economy	GDP per capital, R&D expenditure in % of GDP, Unemployment rate, Gini coefficient
Smart energy	Energy efficiency, Clean energy Utilization, Energy storage, Energy cost
Smart governance	Government Effectiveness, Government Digitalization, Public Participation, Corruption Rate
Smart infrastructure	Infrastructure investment, Sustainable infrastructure, Smart device penetration, Water resources
Smart environment	Air quality, Climate change—GHG emissions, Waste management, Biodiversity & Habitat
Smart society	Educational level, Poverty rate, Gender equality, Healthcare index
Pandemic resiliency	Response rate, Death toll, Confirmed cases, Infrastructure capacity
Smart transportation	Transport efficiency, Technology Integration, Traffic Congestion, Accessibility

The UN-Habitat Global Urban Monitoring Framework (UMF) [37] provides a comprehensive structure for monitoring urban development. The framework is consisted of 5 dimensions. And each dimension is linked with 4 objectives such as being safe and peaceful, inclusive, resilient and sustainability. This framework is designed for monitoring urban development rather than producing a score. Dimensions and indicators of UMF are provided in Table 17.

Table 17. Dimensions and Indicators of UMF [37].

Dimensions	Indicators
Society	Under-5 mortality rate, Basic services, Life expectancy at birth, Slum population, among others.
Economy	Children engaged in child labor, Unemployment rate, City product (GDP) per Capita (PPP), Sub-national debt, among others.
Environment	Wastewater safely treated, Access to open public spaces, Renewable energy share, Total greenhouse gas emissions per year/per capita, among others.
Culture	Culture for social cohesion, Access to culture, Cultural employment, Sustainable management of heritage, among others.
Governance and Implementation	Victims of intentional homicide, Participation in urban planning and management, Own revenue collection, Registered births, among others.

Note: Each dimension of UMF [37] includes several sub-indicators. Only representative examples are shown for brevity.

The UN-Habitat Global Urban Monitoring Framework [37] also outlines a dimension-based structure for urban assessment. Similar to the proposed SSC framework, it adopts a multidimensional perspective and relies on standardized indicators. However, the UMF mainly provides general guidance for monitoring purposes and does not specify a detailed methodology for normalization, weighting, and aggregation. In contrast, the proposed framework defines these steps explicitly to ensure consistent and comparable results across countries.

The indices and their methodologies identified in the literature review are summarized and compared in Table 18.

Table 18. Methodological Comparison of Existing Smart and Sustainable City Indices.

Source	Normalization	Weighting	Aggregation & Scoring
IESE Cities in Motion Index [24]	Not specified	Not specified	Not specified
Arcadis Sustainable Cities Index [25]	Min-Max	Equal	Arithmetic mean
Corporate Knights Sustainable Cities Index [26]	Min-Max	Expert evaluation	Weighted sum
Sustainable Development Report Index [27]	Min-Max	Equal	Arithmetic mean
City Prosperity Index [28]	Min-Max	Equal	Arithmetic mean
IMD Smart City Index [29]	Min-Max	Equal	Arithmetic mean
GPCI Index [31]	Not specified	Not specified	Not specified
India Smart City Index [32]	Decile-based	Equal	Arithmetic mean
Tas and Alptekin [33]	MEREC	MEREC	MARCOS
Milad Pira [20]	No methodological framework proposed
Gazzeh [34]	No methodological framework proposed
Ozkaya and Erdin [21]	TOPSIS	ANP	TOPSIS
Shmelev and Shmeleva [22]	Z-score standardization	Equal	Weighted sum
Haksevenler et al. [18]	Min-max	Equal	Weighted sum
Cohen [15]	No methodological framework proposed
Abu-Rayash and Dincer [5]	0–1 Normalization	Various weighting schemes	Weighted sum
Urban Monitoring Framework [36]	Min–Max	Various weighting	Arithmetic mean
Proposed Methodological Framework	Goal-Post Based Normalization	Redundancy-Penalized Entropy Weighting	Soft Non-Compensatory Aggregation

The review of existing smart city and sustainable city indices shows that the integrated evaluation of smartness and sustainability remains limited in the current literature. In addition, many indices rely on simplified methodological choices, such as min–max normalization, equal weighting, and the calculation of overall scores through simple arithmetic averaging. These observations indicate a clear need for a more comprehensive and transparent SSC assessment approach. Motivated by this gap, this study aims to develop a composite SSC index and a systematical methodological framework that enables consistent scoring and comparison, while addressing both conceptual and methodological limitations observed in existing index-based studies.

From a broader perspective, urban indicator systems have evolved over time toward more integrated and multidimensional approaches. According to Gomez-Alvarez et al. [23], recent developments can be characterized as a “third generation” of urban indicator systems, which emphasize integrated, multidimensional, and policy-relevant frameworks. In line with this perspective, the proposed SSC index can be positioned within this third generation, as it brings together multiple dimensions and integrates smart city and sustainability aspects within a single structure [23]. In addition, Wong [38] highlights several key principles for developing robust and globally relevant urban indicator systems, such as consistency, comparability, and policy relevance. The proposed SSC framework follows these principles by relying on standardized international data sources, ensuring comparability across countries, and applying a transparent evaluation process. At the same time, certain aspects, such as the use of qualitative information, are not explicitly covered, as the focus of this study is on quantitative and indicator-based assessment.

3. Materials and Methods

3.1. Indicator Selection and Data Collection Process for the SSC Index

The selection of indicators is a critical step in index construction. In this study, the indicator selection process follows a structured procedure. First, existing indices including both academic studies and industry-based index reports are reviewed. Then, based on this review, preliminary pool of candidate indicators is compiled.

Next, a panel of experts evaluated the candidate indicators. The expert group consisted of three specialists with backgrounds in urban planning, sustainability, and smart city technologies. They are selected based on their academic and professional experience. The experts were asked to assess the relevance, clarity, and applicability of each indicator. Based on their evaluations, redundant or conceptually inconsistent indicators were removed.

The seven dimensions used in this study were identified through a synthesis of commonly used dimensions in existing smart city and sustainable city assessment frameworks presented in the literature review. In the proposed framework, the index is formed from indicators. Thus, normalization and weighting methods are applied directly to the indicators. After these procedures, the overall index scores are determined by the normalized indicator values using indicator weights. Therefore, alternative dimension structures do not affect the calculation of the overall index scores. However, they may influence the interpretation of the results, particularly at the dimension level, as well as the way indicators are grouped and presented within the framework. After the screening step, data are collected from internationally recognized databases such as the World Bank, OECD, ITU, UN and international databases to ensure reliability [39,40,41,42,43,44,45,46,47,48,49,50,51,52]. Finally, a data availability assessment is conducted. Indicators for which reliable data could not be obtained are excluded from the empirical application. The overall indicator selection and data collection procedure is illustrated in Figure 1.

Figure 1. Indicator Selection and Data Collection Process.

3.2. Proposed SSC Indexing Model

This study proposes a composite SSC scoring model to assess cities across multiple dimensions of smartness and sustainability. Since the indicators differ in their measurement scales, directions (benefit or cost), and data distributions, the analysis follows a clear, step-by-step procedure that includes normalization, indicator weighting, final aggregation for scoring and robustness analysis under different parameter assumptions. The framework is structured to make indicators comparable, reduce the effect of extreme values and overlapping information, and avoid excessive compensation between dimensions, resulting in a more balanced and conceptually coherent evaluation of urban smart sustainability performance. The proposed SSC Indexing Model is illustrated in Figure 2.

Figure 2. Proposed Sustainable Smart City Indexing Model.

3.2.1. Normalization and Winsorization

The indicators included in the SSC Index are measured in different units and exhibit heterogeneous distributions across cities. As discussed in Wong [38], the indicators could be benefit or cost-oriented. However, in this study, to ensure comparability a goalpost-based normalization approach is employed. Before the normalization process, a winsorization procedure is applied as a conservative data pre-processing step to limit the influence of extreme values.

For each indicator $j$, lower and upper bounds are defined using the 5th and 95th percentiles of its empirical distribution across all cities. Specifically, the lower bound $L_j$ and the upper bound $U_j$ are computed as:

$$L_j=Q_{0.05}\left({x.}_j\right),U_j=Q_{0.95}\left({x.}_j\right)$$

(1)

where $Q_{0.05}\left({x.}_j\right)$ and $Q_{0.95}\left({x.}_j\right)$ denote the 5th and 95th percentiles of indicator $j$ calculated across all cities. Observed values that fall outside this range are adjusted using the following winsorization rule:

$$x_{ij}^{\prime }=\left\{\begin{array}{cc}{L}_j& \mathrm{if}{x}_{ij}<L_j\\ x_{ij}& \mathrm{if}{L}_j\le x_{ij}\le U_j\\ U_j& \mathrm{if}{x}_{ij}>U_j\end{array}\right.$$

(2)

where $x_{ij}$ represents the observed value of indicator $j$ for city $i$, and $x_{ij}^{\prime }$ denotes the winsorized value. The 5th and 95th percentile thresholds are applied to reduce the effect of extreme values without removing too much information from the dataset. This allows the main pattern of the data to remain intact while limiting the influence of outliers. The adjusted indicator values are then normalized to a unit-free scale ranging from 0 to 1. Indicators where higher values represent better performance (benefit-oriented) and those where lower values indicate better performance (cost-oriented) are normalized using, respectively, Formulations (3) and (4):

$$n_{ij}={\displaystyle \frac{x_{ij}^{\prime }-L_j}{U_j-L_j}}$$

(3)

$$n_{ij}={\displaystyle \frac{U_j-x_{ij}^{\prime }}{U_j-L_j}}$$

(4)

where $n_{ij}$ denotes the normalized value of indicator j for city i obtained after the winsorization and goalpost-based normalization procedures. As a result, all normalized indicator values are expressed on a common scale, where higher values represent better performance across all indicators.

3.2.2. Indicator Weighting: Redundancy-Penalized Entropy Weighting (RPEW)

To determine indicator weights objectively while accounting for information overlap, a Redundancy-Penalized Entropy Weighting (RPEW) approach is adopted. The method is grounded in information entropy theory, which evaluates the amount of information conveyed by each indicator based on its dispersion across alternatives [53]. To further address redundancy among indicators, an additional correlation-based penalty term is incorporated, ensuring that indicators conveying overlapping information do not receive disproportionate weights [54].

First, a probability matrix is constructed based on the normalized indicator values. For each indicator $j$, the probability associated with city $i$ is computed as:

$$p_{ij}={\displaystyle \frac{n_{ij}}{\sum _{i=1}^m{n}_{ij}}}$$

(5)

where $m$ denotes the total number of cities, ensuring that, for each indicator $j$, the probabilities $p_{ij}$ sum to unity across all cities.

Next, the entropy value of each indicator $j$ is then computed as:

$$e_j=-{\displaystyle \frac{1}{\mathrm{l}\mathrm{n}\left(m\right)}}\sum _{i=1}^m{p}_{ij}\mathrm{l}\mathrm{n}\left(p_{ij}\right)$$

(6)

where $1/\mathrm{l}\mathrm{n}\left(m\right)$ is a normalization constant ensuring that $0\le e_j\le 1.$ For observations with $p_{ij}=0$, the term $p_{ij}\ln \left(p_{ij}\right)$ is treated as zero, consistent with the limit properties of the entropy function.

The degree of divergence, representing the discriminatory power of each indicator, is defined using following formula:

$$d_j=1-e_j$$

(7)

To reduce redundancy among indicators, a correlation-based penalty factor is applied. The Pearson correlation coefficient between indicators $j$ and $k$ denoted by ${\rho }_{jk}$. This coefficient is calculated using the normalized values of the indicator across all cities. Based on this, the redundancy penalty is defined as follows:

$$R_j=1-{\displaystyle \frac{1}{K-1}}\sum _{k\ne j}\mid {\rho }_{jk}\mid$$

(8)

where $K$ is the total number of indicators. This formulation penalizes indicators that convey information already captured by others, regardless of correlation direction.

The final indicator weights are obtained by combining both components:

$$w_j={\displaystyle \frac{d_j\cdot R_j}{\sum _j{d}_j\cdot R_j}}$$

(9)

This procedure ensures that indicators with higher discriminatory power and lower redundancy receive greater importance in the index construction.

3.2.3. Aggregation

The proposed SSC Index Model adopts a two-stage aggregation structure. It proceeds from indicators to dimensions and subsequently to an overall city score.

Each indicator $j$ is assigned to a predefined dimension (e.g., economy, environment, governance, mobility). To prevent dimensions with a larger number of indicators from dominating the results, indicator weights are re-normalized within each dimension:

$${\tilde{w}}_{j\mid d}={\displaystyle \frac{w_j}{\sum _{k\in J_d}{w}_k}}$$

(10)

where $J_d$ denotes the set of indicators belonging to dimension $d$.

The dimension score for city $i$ is then calculated as:

$$S_{i,d}=\sum _{j\in J_d}{\tilde{w}}_{j\mid d}{n}_{ij}$$

(11)

This weighted aggregation allows compensability among indicators within the same dimension, reflecting their complementary nature.

3.2.4. Overall SSC Scoring

To calculate the overall SSC score, the study uses a soft non-compensatory aggregation based on the geometric mean. Unlike the arithmetic mean, this approach does not allow very strong performance in one dimension to fully make up for weak performance in another. This is in line with the multidimensional character of smart and sustainable urban development, where balanced progress across dimensions is more meaningful than excellence in only a few areas.

The overall SSC score for city $i$ is defined as:

$$SS{C}_i=\mathrm{e}\mathrm{x}\mathrm{p}\left(\sum _{d=1}^D{W}_d\mathrm{l}\mathrm{n}(S_{i,d}+\epsilon )\right)$$

(12)

where $D$ is the number of dimensions, $W_d$ denotes dimension weights, and $\epsilon$ is a small constant set to $\epsilon ={10}^{-6}$ introduced to avoid numerical issues when dimension scores approach zero.

This scoring strategy favors cities that perform consistently across all dimensions, while disadvantaging those with strong imbalances between dimensions. In this way, the index remains closely aligned with the core idea of smart and sustainable cities, which emphasizes balanced and integrated urban development.

3.2.5. Robustness Analysis

To evaluate the robustness of the proposed SSC Indexing Model, the study conducts an analysis. This analysis is applied under equal weighting and conventional entropy-based weighting assumptions. The empirical results are presented and discussed in Section 4 and Section 5, respectively.

4. Application of the Proposed SSC Indexing Model

4.1. SSC Index Structure and Indicator Set

The case study is built around the proposed SSC Index, a multidimensional composite framework developed to evaluate smartness and sustainability in urban systems in an integrated way. Rather than treating these two concepts separately, the index is designed to reflect how economic performance, digital innovation, environmental sustainability, governance quality, mobility systems, social well-being, and urban form interact with one another. In its final version, the SSC Index consists of seven dimensions and 55 indicators, each clearly specified in terms of measurement unit, benefit–cost orientation, and data source.

The Economic Performance and Competitiveness dimension captures the overall economic strength of cities, together with labor market conditions, income distribution, and institutional factors that shape urban economic outcomes. Indicators in this dimension reflect productivity levels, growth dynamics, employment conditions, investment attractiveness, and the effectiveness of regulatory and institutional frameworks.

The Innovation and Digital dimension represents a city’s technological capacity and its ability to generate, adopt, and diffuse knowledge. It includes indicators related to research and development activity, the availability of skilled human capital in science and technology, digital connectivity, the maturity of e-government services, and access to open data.

The Environment dimension focuses on environmental quality, resource efficiency, and climate resilience. It brings together indicators related to greenhouse gas emissions, renewable energy use, air quality, waste and water management practices, ecosystem protection, and exposure to environmental pressures.

The Mobility dimension addresses the performance of urban transport systems from the perspective of efficiency, sustainability, and safety. Indicators in this dimension cover traffic safety, transport-related emissions, vehicle ownership patterns, electrification, public transport usage, logistics infrastructure, accessibility, and rail network density.

The Governance dimension reflects the quality of institutions and public administration at the urban level. It includes indicators related to government effectiveness, transparency and accountability, control of corruption, civic participation, disaster preparedness, digital public service provision, and fiscal openness.

The Human Development and Social Inclusion dimension evaluates social well-being, equity, and access to essential services. It integrates indicators on health outcomes, educational attainment, youth participation in the labor market, social protection mechanisms, gender equality, access to sanitation, and broader social stability.

Finally, the Urban Form and Livability dimension captures the physical structure of cities and the quality of everyday urban life. Indicators in this dimension reflect urbanization patterns, population dynamics, housing affordability, access to green spaces, transport accessibility, and the overall quality of urban infrastructure.

Table 19 summarizes the full structure of the proposed SSC index, covering all seven dimensions and the 55 indicators that form its conceptual and methodological basis. This complete set of indicators defines the intended scope of the SSC index, independent of the data limitations encountered in the case study. For each indicator, the table reports its definition, measurement unit, benefit–cost orientation, and data source, ensuring transparency and supporting the reproducibility of the index.

Table 19. Full Structure of the Proposed Sustainable Smart City Index.

Dimension	Code	Indicator	Unit	Description	Source	Benefit/Cost
Economic Performance and Competitiveness	ECO1	GDP per capita (PPP)	USD (PPP)	Economic output per person adjusted for purchasing power.	World Bank [39]	Benefit
Economic Performance and Competitiveness	ECO2	GDP growth rate	%	Annual percentage growth of real GDP.	World Bank [39]	Benefit
Economic Performance and Competitiveness	ECO3	Unemployment rate	%	Share of labor force without employment.	World Bank [39]	Cost
Economic Performance and Competitiveness	ECO4	Labor force participation	%	Population aged 15–64 participating in labor market.	World Bank [39]	Benefit
Economic Performance and Competitiveness	ECO5	FDI inflow per capita	USD/person	Net foreign direct investment per capita.	World Bank [39]	Benefit
Economic Performance and Competitiveness	ECO6	High-tech exports share	% of exports	Share of high-tech products in total exports.	World Bank [39]	Benefit
Economic Performance and Competitiveness	ECO7	Regulatory quality score	Index	Institutional quality regarding regulations.	WGI [40]	Benefit
Economic Performance and Competitiveness	ECO8	Gini coefficient	0–1	Income inequality measure (lower is better).	OECD [41]	Cost
Innovation & Digital	INN1	R&D expenditure	% of GDP	Share of national output spent on research & development.	World Bank [39]	Benefit
Innovation & Digital	INN2	Researchers per million	Researchers/million	Researchers engaged in R&D per million people.	World Bank [39]	Benefit
Innovation & Digital	INN3	Patent applications	Number of applications	Number of patent applications filed by residents	World Bank [39]	Benefit
Innovation & Digital	INN4	Internet usage	% population	Individuals using the internet.	World Bank [39]	Benefit
Innovation & Digital	INN5	Mobile broadband subscriptions	Subscriptions/100 people	Active mobile broadband subscriptions.	ITU [42]	Benefit
Innovation & Digital	INN6	Fixed broadband subscriptions	Subscriptions/100 people	Fixed broadband penetration.	ITU [42]	Benefit
Innovation & Digital	INN7	E-Government Development Index	Index	UN assessment of digital government maturity.	UN E-Government Knowledgebase [43]	Benefit
Innovation & Digital	INN8	Open Data Index	Index	Government open data readiness and availability.	Not Available for empirical application	Benefit
Environment	ENV1	CO2 emissions per capita	ton/person	Territorial CO2 emissions divided by population.	OWID [44]	Cost
Environment	ENV2	Renewable energy consumption	%	Share of renewables in total energy use.	IEA [45]	Benefit
Environment	ENV3	PM2.5 concentration	µg/m3	Population-weighted annual average PM2.5.	World Bank [39]	Cost
Environment	ENV4	Municipal solid waste per capita	kg/person/year	Annual solid waste generated per person.	Not Available for empirical application	Cost
Environment	ENV5	Wastewater treatment coverage	% population	Share of population connected to wastewater treatment.	Not Available for empirical application	Benefit
Environment	ENV6	Access to safe drinking water	% population	People using safely managed drinking water services.	Not Available for empirical application	Benefit
Environment	ENV7	Forest area ratio	% land area	Proportion of land area covered by forest.	The Global Economy [46]	Benefit
Environment	ENV8	Climate resilience index	Index	Composite indicator of climate vulnerability/resilience.	ND-GAIN [47]	Benefit
Environment	ENV9	Water stress level index	Index	Level of national water stress based on a 0–5 scale.	World Population Review [48]	Cost
Mobility	MOB1	Traffic fatality rate	Deaths/100,000 people	Road traffic deaths per 100,000 population.	WHO [49]	Cost
Mobility	MOB2	Transport CO2 emissions per capita	kg CO2 per capita	Carbon dioxide emissions generated by the transport sector per person	World Bank [39]	Cost
Mobility	MOB3	Passenger car ownership	Vehicles per 1000 people	Number of registered passenger cars per 1000 people.	Not Available for empirical application	Cost
Mobility	MOB4	Electric vehicle adoption	% of new vehicle sales	Share of electric vehicles in total new passenger car sales.	IEA [45]	Benefit
Mobility	MOB5	Public transport modal share	% of total trips	Proportion of trips made using public transport.	Not Available for empirical application	Benefit
Mobility	MOB6	Air transport passengers per capita	Passengers per capita	Number of air transport passengers.	Not Available for empirical application	Benefit
Mobility	MOB7	Logistics Performance Index (Infrastructure score)	Index	Assessment of the quality of transport and logistics infrastructure.	World Bank [39]	Benefit
Mobility	MOB8	Access to public transport	% of population	Share of population with convenient access to public transport services.	Not Available for empirical application	Benefit
Mobility	MOB9	Rail lines density	km	National railway density.	Not Available for empirical application	Benefit
Governance	GOV1	Government effectiveness	Index	Perception of public service quality and governance.	WGI [40]	Benefit
Governance	GOV2	Control of corruption	Index	Control of corruption within public institutions.	WGI [40]	Benefit
Governance	GOV3	Voice & accountability	Index	Citizens’ participation in government.	WGI [40]	Benefit
Governance	GOV4	E-Participation Index	Index	Digital participatory services offered by government.	UN E-Government Knowledgebase [43]	Benefit
Governance	GOV5	Local digital services coverage	0–1	Extent of online municipal public services.	Not Available for empirical application	Benefit
Governance	GOV6	Disaster preparedness score	Index	Readiness and risk reduction capacity.	Not Available for empirical application	Benefit
Governance	GOV7	Budget transparency	Index	Openness of public budget processes.	Not Available for empirical application	Benefit
Social Inclusion and Human Development	SOC1	Life expectancy	Years	Average expected lifespan at birth.	World Bank [39]	Benefit
Social Inclusion and Human Development	SOC2	Infant mortality	Deaths/1000 births	Infant deaths per 1000 live births.	World Bank [39]	Cost
Social Inclusion and Human Development	SOC3	Education attainment	Years	Average years of formal education for individuals aged 15–64.	OWID [44]	Benefit
Social Inclusion and Human Development	SOC4	Youth unemployment	%	Unemployment rate among youth aged 15–24.	International Labour Organization [50]	Cost
Social Inclusion and Human Development	SOC5	Social protection coverage	%	Population covered by social protection systems (at least one benefit)	International Labour Organization [50]	Benefit
Social Inclusion and Human Development	SOC6	Gender inequality index	Index	Gender inequality.	International Labour Organization [50]	Cost
Social Inclusion and Human Development	SOC7	Access to sanitation	% population	Safely managed sanitation services.	UNICEF JMP [51]	Benefit
Social Inclusion and Human Development	SOC8	Peace index	Index	National peacefulness based on societal safety, conflict levels, and militarisation.	Institute for Economics and Peace [52]	Cost
Urban Form and Livability	URB1	Urban population share	%	Percentage of population living in urban areas.	World Bank [39]	Benefit
Urban Form and Livability	URB2	Urban population growth	%	Annual growth rate of urban population.	World Bank [39]	Cost
Urban Form and Livability	URB3	Housing affordability (price-income ratio)	Index	Ratio of housing prices to income.	Not Available for empirical application	Cost
Urban Form and Livability	URB4	Urban green space availability	m2/person	Public green area per capita.	Not Available for empirical application	Benefit
Urban Form and Livability	URB5	Transit station density	Stations/km2	Density of major transit stations.	Not Available for empirical application	Benefit
Urban Form and Livability	URB6	Infrastructure quality index	Index	Quality of national infrastructure systems.	World Population Review [48]	Benefit

4.2. Scope and Data Selection

The empirical application of the proposed SSC index is conducted within a clearly defined and comparable analytical scope. While the SSC methodology is conceptually designed for city-level assessment, the case study makes use of country-level data because comparable city-level information is not sufficiently available across countries. All indicators used in the analysis are based on officially reported and verifiable data obtained from internationally recognized sources, as detailed in Table 19.

The dataset used in this study was compiled from publicly available international statistical databases, including the World Bank Open Data, OECD Data, ITU Statistics, UN E-Government Knowledgebase, Our World in Data (OWID), ND-GAIN database, World Governance Indicators (WGI), WHO, ILO, UNICEF JMP, Institute for Economics and Peace, and The Global Economy database [39,40,41,42,43,44,45,46,47,48,49,50,51,52]. The corresponding source of each indicator is reported in Table 19.

The case study focuses on 38 OECD countries, which are selected to ensure a reasonable level of institutional, economic, and statistical comparability. OECD countries generally follow similar data quality standards and reporting practices, and they offer relatively consistent time series. This makes them suitable for cross-country benchmarking within the context of smart and sustainable cities. The deliberate focus on this group also helps avoid arbitrary country selection and strengthens the internal consistency of the case study.

The year 2023 is chosen as the reference period for the empirical application. While more recent data are not yet fully available across all indicators and countries, 2023 represents the most recent year for which a sufficiently complete and comparable dataset can be assembled.

Although the proposed SSC index is theoretically defined by 55 indicators, limitations in data availability require a smaller indicator set for the empirical analysis. In particular, 15 indicators are excluded from the case study because comparable cross-country data are not available. These exclusions are based solely on data constraints and do not reflect conceptual or methodological choices. As a result, the case study is conducted using a final set of 40 indicators.

Minor data gaps remain for certain indicators and countries in the 2023 dataset. When a limited number of indicator values are missing for a given country, these values are imputed using the mean of the corresponding indicator across the remaining OECD countries. In cases where 2023 data are entirely unavailable but historical observations exist; missing values are estimated using a machine learning–based polynomial forecasting approach. This forecasting step is applied conservatively and solely to ensure continuity of the dataset; methodological details are intentionally kept concise, as forecasting is not the primary focus of the study.

Through these data selection and preparation procedures, the case study demonstrates the applicability of the proposed SSC methodology under real-world data constraints while maintaining transparency, consistency, and analytical rigor. The resulting dataset forms the basis for the SSC score computation and country ranking presented in the following sections.

4.3. Indicator Screening and Data Treatment

While the proposed SSC index comprises 55 indicators in its full structure, the empirical implementation of the case study requires a systematic screening and treatment of indicators based on data availability conditions. As a result of this process, indicator exclusion and data forecasting are applied to adjust the available data.

First, 15 indicators are excluded from the case study due to the absence of consistent and comparable cross-country data for the selected reference year. The excluded indicators are INN8, ENV4, ENV5, ENV6, MOB3, MOB5, MOB6, MOB8, MOB9, GOV5, GOV6, GOV7, URB3, URB4, and URB5. These indicators are removed solely because reliable data could not be obtained. Importantly, this exclusion is driven by empirical data constraints rather than conceptual or methodological considerations.

Second, for seven indicators, complete data for 2023 are not available for all countries, even though historical observations exist. These indicators—INN1, INN2, INN3, ENV2, ENV3, MOB1, and MOB2—are therefore kept in the case study, and the missing 2023 values are estimated using a machine learning–based polynomial forecasting approach. This forecasting step is applied in a cautious manner, with the aim of maintaining temporal continuity while preserving the overall structure of the data.

After excluding indicators with unavailable data and addressing partially missing values through forecasting, the final case study dataset includes 40 indicators covering all seven dimensions of the proposed SSC index. This approach allows the analysis to remain empirically feasible while preserving the multidimensional structure and overall coherence of the proposed SSC framework.

4.4. SSC Scoring and Country Rankings

Following the data preparation and indicator screening procedures described in the previous sections, final SSC scores are calculated for each country using the proposed SSC Indexing Model. The empirical analysis is based on 40 indicators across seven dimensions and covers 38 OECD countries for the reference year 2023.

For each country, normalized indicator values are combined with the corresponding RPEW-based weights to obtain a single composite SSC score. Higher scores indicate stronger and more balanced performance in terms of sustainability and smartness. Based on these scores, countries are ranked in descending order in Table 20.

Table 20. SSC Scores and Rankings of OECD Countries (2023).

Countries	Scores	Rankings
Norway	0.68659	1
Denmark	0.68433	2
Sweden	0.68128	3
Iceland	0.66302	4
Finland	0.66217	5
Japan	0.65304	6
Switzerland	0.63066	7
Republic of South Korea	0.62663	8
Netherlands	0.59496	9
Germany	0.58296	10
United Kingdom	0.53882	11
New Zealand	0.53172	12
United States	0.51911	13
Estonia	0.51217	14
Austria	0.51199	15
France	0.49305	16
Ireland	0.48738	17
Australia	0.48455	18
Luxembourg	0.47896	19
Belgium	0.46571	20
Canada	0.46258	21
Slovenia	0.44141	22
Latvia	0.44055	23
Lithuania	0.42553	24
Israel	0.41554	25
Spain	0.40457	26
Portugal	0.39707	27
Poland	0.38433	28
Czech Republic	0.37281	29
Italy	0.35512	30
Greece	0.35108	31
Slovak Republic	0.32105	32
Hungary	0.30194	33
Chile	0.29485	34
Costa Rica	0.25742	35
Turkiye	0.21378	36
Mexico	0.17948	37
Colombia	0.14622	38

The results reveal a clear differentiation in SSC performance levels across the sample. Countries such as Norway, Denmark, Sweden, Iceland, and Finland appear at the top of the ranking, whereas relatively lower SSC scores such as Türkiye, Mexico and Colombia are observed for countries positioned at the lower end of the distribution.

While Table 20 presents the overall scores, examining dimension-level results provides additional insight into the drivers of cross-country differences. Table 21 presents the ranking of countries withing each SSC dimension. The results show that having a strong performance in one dimension is not enough to appear at the top of the ranking. For example, the United States ranks first in the Innovation & Digital dimension. However, it ranks considerably lower in the overall SSC index due to relatively weaker performance in other dimensions such as mobility and environmental dimensions. Another notable example is Germany. It is ranked fourth in both Innovation & Digital and Social Inclusion & Human Development dimensions. However, its relatively moderate performance in other dimensions such as mobility and environmental indicators limits its overall SSC ranking. Estonia has the same patters as Germany and United States. Although it ranks fourth in the Environment dimension, lower scores in the other dimensions place Estonia in the middle of the overall SSC ranking. The dimension-level results also shows that countries ranked at the top of the overall SSC index tend to perform relatively well across several dimensions rather than excelling in only one area. In other words, balanced performance appears to be an important factor behind higher overall scores. Norway is a good example of this pattern. The country performs strongly in multiple dimensions, including mobility, environment, governance, and social indicators, which helps explain its leading position in the overall SSC ranking. Another notable pattern can be observed when comparing Austria and Canada. Although the two countries have relatively similar performance in several dimensions such as Economic Performance and Competitiveness, Social Inclusion & Human Development, and Urban Form and Livability, Austria achieves considerably higher scores in the environment and mobility dimensions. Despite Canada’s strong governance performance, these differences lead to a noticeable gap in the overall SSC ranking. This example illustrates the non-compensatory nature of the proposed scoring framework, where weak performance in certain dimensions cannot be fully offset by strong results in others.

Table 21. Dimension-Level Rankings of OECD Countries (2023).

Economic Performance and Competitiveness	Innovation & Digital	Environment	Mobility	Governance	Social Inclusion & Human Development	Urban Form and Livability
Iceland	United States	Sweden	Norway	Denmark	Switzerland	Japan
Norway	Republic of South Korea	Finland	Sweden	New Zealand	Japan	Sweden
Australia	Japan	Norway	Finland	Finland	Iceland	Denmark
Switzerland	Germany	Estonia	Iceland	Netherlands	Germany	Finland
Republic of South Korea	Denmark	Latvia	Denmark	Norway	Republic of South Korea	Republic of South Korea
Denmark	United Kingdom	Iceland	Netherlands	Iceland	Australia	Belgium
Ireland	Estonia	Costa Rica	Belgium	Germany	Netherlands	Netherlands
Japan	Finland	Austria	Switzerland	Switzerland	Norway	Germany
United States	Sweden	Switzerland	Luxembourg	Canada	Denmark	United States
Netherlands	Iceland	New Zealand	Austria	Ireland	Slovenia	France
Israel	Netherlands	Denmark	Ireland	Japan	Ireland	Norway
United Kingdom	Switzerland	Colombia	Portugal	Australia	Canada	Greece
Germany	Norway	Slovenia	New Zealand	United Kingdom	Israel	Chile
Czech Republic	Israel	Portugal	Israel	Estonia	Austria	Switzerland
New Zealand	Australia	Lithuania	Slovenia	Sweden	New Zealand	Costa Rica
Sweden	Austria	Canada	Lithuania	Republic of South Korea	Belgium	United Kingdom
Belgium	France	Japan	Greece	United States	Sweden	Luxembourg
Slovenia	Poland	France	Estonia	Luxembourg	United Kingdom	Italy
Canada	Belgium	Germany	United Kingdom	Austria	Luxembourg	Iceland
Latvia	Spain	Slovak Republic	Germany	France	Finland	Israel
Poland	New Zealand	Luxembourg	France	Lithuania	France	Spain
France	Luxembourg	United Kingdom	Slovak Republic	Chile	Czech Republic	Latvia
Austria	Ireland	Ireland	Australia	Slovenia	Estonia	Australia
Slovak Republic	Canada	Spain	Latvia	Spain	Lithuania	Hungary
Hungary	Lithuania	Czech Republic	Spain	Latvia	Portugal	Mexico
Mexico	Portugal	Hungary	Hungary	Belgium	Italy	Turkiye
Estonia	Slovenia	Republic of South Korea	Italy	Israel	Latvia	New Zealand
Finland	Greece	Poland	Canada	Portugal	Poland	Poland
Lithuania	Italy	Italy	Czech Republic	Costa Rica	Spain	Colombia
Luxembourg	Latvia	Netherlands	Republic of South Korea	Poland	United States	Canada
Portugal	Czech Republic	Greece	Poland	Czech Republic	Greece	Czech Republic
Costa Rica	Hungary	United States	Costa Rica	Italy	Hungary	Austria
Greece	Turkiye	Mexico	Japan	Slovak Republic	Slovak Republic	Lithuania
Italy	Chile	Chile	Turkiye	Turkiye	Chile	Slovak Republic
Spain	Slovak Republic	Australia	United States	Greece	Mexico	Estonia
Chile	Costa Rica	Belgium	Colombia	Colombia	Costa Rica	Ireland
Turkiye	Mexico	Turkiye	Chile	Mexico	Colombia	Slovenia
Colombia	Colombia	Israel	Mexico	Hungary	Turkiye	Portugal

These findings not only show differences in SSC performance across countries but also serve as a foundation for discussing their managerial and policy implications in Section 5.

4.5. Robustness Analysis

Composite indices are often sensitive to methodological assumptions, particularly regarding the weighting of indicators. Different weighting methods may change the relative importance assigned to indicators which will affect the final ranking results. For this reason, robustness analysis is commonly conducted in index studies to examine whether the main findings remain stable under alternative methodological settings [55].

To evaluate the robustness of the proposed RPEW method, an additional analysis is carried out under equal weighting and conventional entropy weighting approaches. In all three scenarios, the same dataset, normalization procedure, and aggregation structure are preserved.

In the equal weighting approach, every indicator has the same importance. This is the case even if some indicators contain more information than others. Conventional entropy weighting partly improves this situation because indicators with greater variation across countries receive higher weights. However, the entropy method does not consider whether some indicators reflect very similar information. When indicators measure closely related aspects, their influence may be counted more than once. The proposed RPEW method tries to limit this problem. It introduces a penalty based on the correlations between indicators. If two indicators are highly related, their effective weights are reduced. In this way, groups of strongly correlated indicators are less likely to dominate the final SSC score.

The country rankings obtained under the three weighting approaches are shown in Table 22.

Table 22. Comparison of Country Rankings under Alternative Weighting Schemes.

Country	Rank (RPEW)	Rank (Equal Weighting)	Rank (Entropy Weighting)
Norway	1	3	1
Denmark	2	1	2
Sweden	3	2	3
Iceland	4	6	5
Finland	5	4	4
Japan	6	8	7
Switzerland	7	5	6
Republic of South Korea	8	11	9
Netherlands	9	7	8
Germany	10	9	10
United Kingdom	11	10	11
New Zealand	12	12	12
United States	13	21	14
Estonia	14	14	15
Austria	15	13	13
France	16	19	19
Ireland	17	17	16
Australia	18	15	17
Luxembourg	19	18	18
Belgium	20	16	20
Canada	21	20	21
Slovenia	22	22	22
Latvia	23	24	23
Lithuania	24	23	25
Israel	25	25	24
Spain	26	26	27
Portugal	27	27	26
Poland	28	29	29
Czech Republic	29	28	28
Italy	30	31	30
Greece	31	30	31
Slovak Republic	32	32	32
Hungary	33	33	33
Chile	34	34	34
Costa Rica	35	35	35
Turkiye	36	36	36
Mexico	37	37	37
Colombia	38	38	38

The analysis shows that overall rankings remain largely stable across three weighting approaches. Countries that have strong performances in seven dimensions such as Norway, Denmark, Sweden, and Finland appear at the top of the ranking. Similarly to proposed model countries that have weak performances in all dimensions, such as Chile, Costa Rica, Turkiye, Mexico and Colombia, these appear at the bottom of the ranking. Results are analyzed deeply in the next section.

5. Discussion and Managerial Implications

The results suggest useful differences in SSC performance across OECD countries. The countries that are placed at the top of the ranking have strong performances in every dimension of the proposed index. Norway, Denmark, Sweden, Iceland, and Finland illustrate this pattern. These countries do not always rank first in every category, but they maintain balanced performance across seven dimensions. Norway is a good example. It ranks highly in mobility, economy, and environment. It also maintains a good position in governance, social inclusion and human development. In contrast to Nordic countries, Colombia, Mexico and Türkiye perform poorly in every aspect which position them at the bottom of the ranking.

Another observation is that strong performance in one dimension alone does not guarantee a high overall ranking. The United States provides a clear example. Although it ranks first in the Innovation and Digital dimension, its overall SSC position is much lower. This suggests that technological leadership alone is not sufficient if other areas such as environment, governance, mobility, or social inclusion perform relatively weaker. Similar patterns can also be seen for Ireland and Belgium. Ireland is strong economically among others, yet its performance in urban form and livability dimension is weaker. Belgium performs better then Ireland in urban form and livability dimension but it has weak score in governance dimension. These countries perform strongly in some dimensions, yet they remain in the middle of the ranking due to weaker results in others. In this sense, the SSC framework highlights the importance of balanced development rather than isolated strengths.

The robustness analysis shows that the SSC rankings are generally stable across different weighting methods. When equal weighting and conventional entropy weighting are applied, the overall ranking structure does not change significantly. Countries that perform very well or very poorly tend to keep similar positions.

However, some moderate shifts can be observed for countries with uneven performance across dimensions. For example, the position of the United States and Australia declines under equal weighting, while countries such as Republic of South Korea and France move slightly upward. This indicates that weighting assumptions mainly affect countries whose strengths are concentrated in a limited number of dimensions.

The results also support the use of the RPEW method. Some indicators—especially those related to governance—measure similar aspects of institutional quality and may contain overlapping information. By penalizing highly correlated indicators, RPEW reduces this redundancy and helps ensure that the final SSC scores reflect multidimensional performance.

While the results highlight the importance of balanced performance across SSC dimensions, these dimensions should not be interpreted as fully independent. In practice, many of them influence each other. For example, improvements in governance may support innovation and digital development, while better mobility systems can contribute to environmental performance. In addition, these relationships may vary across countries and may also change over time. Policy makers can use the SSC index as a tool to identify strengths and weaknesses rather than as a fixed measure of performance.

The findings also have practical implications for policymakers and public administrators. The proposed SSC index can be used as a tool to identify areas that require policy attention instead of focusing only on overall rankings. Looking at the results at the dimension and indicator levels makes it possible to see where performance is relatively weak, such as governance, digital infrastructure, environmental management, or innovation capacity. This helps decision-makers focus on specific policy areas rather than applying general policy measures. For instance, if a country performs relatively well in environmental indicators but receives lower scores in governance or technology-related dimensions, policy efforts could be directed toward improving digital infrastructure, strengthening institutional coordination, or supporting innovation activities.

The framework can also be useful for benchmarking and long-term planning. Governments and urban administrations can compare their results with those of other countries. This helps them by observing where they stand in different dimensions of smart and sustainable development. Such comparisons may help identify policy areas that need improvement and allow decision-makers to learn from countries that perform better in certain areas. In this sense, the proposed index may provide a structured way for policymakers to monitor progress and guide future policy decisions related to smart and sustainable development [56].

6. Conclusions

This study aimed to develop a systematic framework for assessing SSC performance by combining smart and sustainable dimensions within a single composite index. In response to the first research question, the findings show that these dimensions can be combined without losing their individual analytical meaning when they are structured within a clear and coherent framework. Indicators are organized under well-defined dimensions. This allows smartness and sustainability to be treated as complementary elements rather than isolated or competing concepts within a consistent scoring logic.

Regarding the second research question, the results suggest that a structured and transparent scoring framework improves the comparability of SSC performance across countries. The proposed methodology highlights cross-country differences more clearly by reducing the influence of scale effects, overlapping indicators, and excessive compensation between dimensions. In addition, the robustness analysis indicates that the overall ranking structure remains largely stable under alternative weighting schemes. As a result, the SSC scores provide a more interpretable and balanced representation of overall urban performance.

The empirical results show that overall SSC rankings are closely related to how balanced countries perform across different dimensions. Countries that achieve relatively strong results in several SSC dimensions tend to rank higher overall. For instance, Nordic countries such as Norway and Sweden have higher performances across environmental, governance, mobility, and social indicators, which explains their leading positions in the SSC ranking. By contrast, some countries such as United States, Ireland and Belgium have high scores in certain dimensions but obtain lower overall scores because their performance is weaker in other areas. This pattern highlights the multidimensional nature of the SSC framework and suggests that strong results in a single dimension alone are not sufficient to have high overall SSC performance.

Despite these contributions, the study is subject to several limitations. First, although the SSC framework is conceptually designed for city-level analysis, country-level data are used due to the limited availability of city-level datasets. Second, data availability constraints required the exclusion of certain indicators from the empirical analysis, even though they remain part of the theoretical SSC framework. Thirdly, while forecasting techniques were applied to address missing recent data, these estimates represent an approximation and may not fully capture short-term structural changes. In addition, the results of the SSC index may be sensitive to variations in data availability, indicator selection, and weighting assumptions, particularly when the framework is applied in contexts with incomplete or heterogeneous datasets. Therefore, users of the model should interpret the results as a comparative analytical tool rather than an absolute measure of urban performance.

These limitations also point to promising directions for future research. In particular, future work will explore the application of the Choquet integral to model interaction effects and interdependencies among indicators and dimensions, which cannot be fully captured by conventional aggregation approaches. In addition, the study plans to employ Network Data Envelopment Analysis to examine SSC performance from a multi-stage efficiency perspective. Future studies may apply the proposed framework to city-level data as more consistent datasets become available, allowing for more detailed and policy-oriented SSC evaluations. Future research could also explore how stakeholder preferences, including those of policy makers, urban planners, and local communities, influence the relative importance of SSC dimensions and indicators within the index framework. Additionally, in future studies, the proposed framework may be systematically compared with other index approaches using the same data.