Content area
With the rapid expansion of electric vehicles (EVs), optimizing charging infrastructure has become essential for advancing sustainable transportation. The integration of Geographic Information Systems (GIS) and Multi-Criteria Decision-Making (MCDM) methods provides a robust scientific framework for selecting electric vehicle charging station (EVCS) locations. This bibliometric study analyses 1336 WoS Core Collection records (2016–2025; 2025 partial-year) and highlights four thematic clusters. Annual output peaks in 2022 (18.039%) and 2024 (18.563%), with leading national shares from China (18.713%), India (16.692%), and Iran (13.323%). VOSviewer network mapping delineates four cohesive clusters—EVCS layout/planning, GIS–MCDM methods, spatial sustainability, and uncertainty/risk—providing a structured lens for evidence-based synthesis. A focused mapping restricted to GIS and MCDM for EVCS siting—read directly from VOSviewer co-occurrence structures and clusters—yields EVCS-specific evidence that domain-wide surveys do not resolve with comparable clarity. The analysis includes keyword co-occurrence, author collaboration networks, country and institutional influence, journal contributions, and citation networks. The study identifies core research themes, key contributors, and high-impact journals in this field. We position the contribution as an EVCS-specific, cluster-wise interpretation of GIS–MCDM developed within a transparent WoS-only pipeline with verbatim query disclosure, which strengthens traceability and is intended to facilitate reproducibility for planning-oriented evidence. The research insights aim to support the development of intelligent and sustainable charging networks.
Introduction
In recent years, Geographic Information Systems (GIS) integrated with Multi-Criteria Decision-Making (MCDM) methods have been extensively applied to Electric Vehicle Charging Station (EVCS) site selection problems, with the aim of optimizing the charging infrastructure layout, promoting electric vehicle adoption, and reducing carbon emissions [1, 2–3]. Existing studies have explored various combinations of GIS-MCDM methods, such as Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), Weighted Aggregated Sum Product Assessment (WASPAS), Analytic Hierarchy Process (AHP), Decision-Making Trial and Evaluation Laboratory (DEMATEL), and MULTIMOORA, to address the complexities of site selection. However, despite advancements, these studies exhibit limitations regarding real-time data integration, renewable energy incorporation, and socioeconomic considerations. This paper reviews recent literature on GIS-MCDM for EVCS site selection, critically analyzes contributions, connections, and limitations, and proposes future research directions. Unlike domain-wide reviews and broad bibliometric mappings [27, 28, 29–30], this study narrows the lens to the combined “GIS + MCDM for EVCS site selection” problem. Within the same VOSviewer pipeline, this topic-specific scope improves the interpretability of the co-occurrence structures for EVCS practice. We therefore position the present analysis as a focused complement to domain-level surveys.
Initially, the application of MCDM methods for EVCS site selection has been extensively studied [1]. employed TOPSIS and WASPAS methods within a GIS environment to assess the suitability of various station locations, highlighting the significance of weighting criteria methods such as CRITIC and Entropy methods in EVCS siting. However, the study neglected the dynamic variations in charging demand, basing decisions primarily on historical data, thereby overlooking critical real-time factors such as traffic flow, grid load, and charging behaviors. Similarly [2], adopted the fuzzy DEMATEL-MULTIMOORA method for optimizing photovoltaic charging station (PVCS) locations, validated by a case study in Qingdao. Nonetheless, their analysis relied predominantly on a predefined set of candidate sites and lacked comprehensive consideration of grid connectivity and user behavioral preferences, potentially impacting the long-term availability of charging stations [4–5].
Multiple studies indicate that MCDM methods are widely adopted in EVCS site selection research [6, 7–8, 6]. were the first to apply fuzzy TOPSIS to evaluate EVCS sites and introduced sustainability indicators, emphasizing the importance of social, environmental, and economic factors. However, their decision criteria were relatively static, neglecting dynamic charging demands [8]. utilized the TOPSIS method integrated with targeted community characteristics to improve the site assessment model [7]. combined UL-MULTIMOORA and DEMATEL, proposing a more adaptive comprehensive evaluation approach and verifying its effectiveness. Nevertheless, these studies predominantly relied on historical data, failing to incorporate real-time traffic flow, grid load, and user charging behaviors, thereby potentially limiting practical applicability. In summary, recent studies indicate a progression from criteria-driven GIS-MCDM (e.g., AHP/TOPSIS) toward objectively weighted schemes (Entropy/CRITIC) and hybrid variants, while the handling of time-varying inputs relevant to EVCS siting remains limited. Accordingly, this research explores trends in GIS-based Multi-Criteria Decision-Making (GIS-MCDM) applications for site selection of EVCS. Specifically, we address the following research question: What have been the development trends of GIS-MCDM applications in the field of EVCS site selection between 2016 and 2025?
Beyond EVCS, cross-domain site-selection studies further substantiate the maturity and transferability of the GIS-MCDM workflow: a fuzzy AHP–TOPSIS design has been employed for highway-restaurant siting, and an Entropy–WASPAS strategy has been applied to campus residence siting; both report stable rankings under sensitivity analysis, offering transferable practices for criteria setting and robustness checks in EVCS research [9]. Within EVCS, a recent fuzzy MCDM model highlights that technological improvement and charging-station expansion are critical to infrastructure effectiveness, which aligns with this study’s methodological focus [10]. Conceptually, EVCS-oriented GIS–MCDM has progressed from axiomatic, criteria-driven schemes such as AHP/TOPSIS [6, 8, 11] to objective-weighting and fuzzy/intuitionistic designs that operationalize uncertainty and preference heterogeneity [7, 12]. Recent studies further blend probabilistic inference and multi-objective optimisation within network and energy-coupling constraints [13–14], shifting the framework from static suitability mapping toward dynamic, constraint-aware decision systems. This trajectory is consistent with domain-level surveys that synthesise methodological evolution and collaboration patterns [15, 16, 17–18].
Optimization models for EVCS site selection
In addition to MCDM methods, optimization models have been employed for EVCS layout optimization [13, 19, 20–21, 20]. proposed a mixed-integer nonlinear programming (MINLP) approach to optimize the location and sizing of fast-charging stations. However, their model primarily emphasized construction costs and grid constraints, insufficiently considering the spatiotemporal distribution of charging demand [21]. presented a spatiotemporal demand coverage optimization method, integrating the electric vehicle rental market into site selection decisions, enhancing station accessibility [19]. used Bayesian networks (BN) for site selection, highlighting the role of fast-charging stations and spatial analysis in GIS environments [13]. further integrated GIS with Bayesian networks to optimize site selection strategies. Despite improvements in model accuracy, most studies still relied on hypothetical data, limiting the generalizability of their outcomes.
With increasing dependence on renewable energy within the EV industry, researchers began exploring PVCS site optimization [3, 22–23, 22]. assessed the economic feasibility and carbon emissions of PVCS, demonstrating their viability, yet neglected the geographical impacts on photovoltaic efficiency [23]. analyzed the match between photovoltaic charging and EV charging demands through spatiotemporal modeling, highlighting research gaps [3]. integrated GIS and MCDM, proposing a systematic PVCS site selection framework, enhancing the rationality of site layouts. However, PVCS research remains largely theoretical, focusing on hypothetical locations, and lacking empirical validation [12]. applied a hybrid fuzzy VIKOR approach to optimize island-based PVCS sites, without considering dynamic load variations and user behaviors.
Furthermore, some studies incorporated renewable energy technologies to optimize EVCS sustainability [3]. proposed a GIS-MCDM framework employing the TODIM method to optimize photovoltaic charging stations in Beijing, primarily considering traffic flow and road distribution without in-depth analysis of photovoltaic power variability effects on site layout [24]. , in Malaysian studies, emphasized scientific planning for photovoltaic electric vehicle charging stations (PEVCS), yet focused mainly on demand analysis without specific site optimization algorithms, limiting the model’s applicability. Comparatively [14], introduced a two-stage multi-objective optimization framework integrating fast-charging stations, photovoltaics, and energy storage systems, simultaneously optimizing infrastructure and renewable energy utilization. However, further empirical research is required to validate its applicability across various urban environments [25].
Furthermore, socioeconomic factors have not been extensively studied in EVCS site selection [26]. , employing the AHP method in Istanbul, assessed economic, environmental, and social criteria influencing EVCS site selection, but mainly relied on subjective weighting and lacked deep analyses of social equity and user behavior [27]. studied EVCS layouts in Winchester, UK, using Google Earth imagery for validation, enhancing practical applicability; however, their model depended heavily on existing road networks and parking facilities, without adequately considering future EV trends affecting site layouts.
In summary, the application of GIS-MCDM for EVCS site selection is well-established, with progress in optimizing transportation infrastructure, integrating renewable energy, and enhancing station accessibility. Nevertheless, existing research faces challenges, including insufficient dynamic data support, renewable energy uncertainties, and limited socioeconomic considerations. Future studies should incorporate machine learning, real-time data analytics, and intelligent optimization algorithms to enhance the flexibility and sustainability of EVCS siting. Additionally, intensified exploration of user behaviors, grid adaptability, and policy factors is necessary to ensure long-term viability and social acceptance of charging infrastructure.
GIS-Based Multi-Criteria Decision-Making (GIS-MCDM)
Recently, the integration of GIS and MCDM has been extensively applied to EVCS site selection, addressing environmental, economic, social and technical influences in location decisions [28, 29–30]. The geographic analysis capability of GIS helps narrow down site selection areas and reduce costs, whereas MCDM methods enable selection among multiple site options. Studies indicate that GIS-MCDM is widely applied in EVCS siting, while ongoing refinement is needed regarding data dynamics, social considerations, and methodological applicability. Recent campus and service-facility studies using Entropy–WASPAS and fuzzy AHP–TOPSIS further validate weighting-and-ranking procedures and provide transferable sensitivity-analysis practices for multi-factor site decisions [9]. An integrated CRITIC–EDAS–CODAS–CoCoSo design shows consistent cross-model rankings under sensitivity checks and offers a tractable template for robustness assessment [31]. Comparatively, AHP/TOPSIS privileges clarity, Entropy/CRITIC improves weight objectivity, and fuzzy variants capture uncertainty; all, however, are sensitive to static inputs—hence the value of periodic updates and sensitivity checks in EVCS applications.
Numerous studies have validated GIS-MCDM’s effectiveness in infrastructure site selection [28]. applied GIS-MCDM to landfill siting, demonstrating methodological reliability. However, landfill site selection primarily considers environmental impacts, while EVCS site selection requires additional dynamic factors such as traffic flow, grid accessibility, and user behaviour [29] and [32]. employed GIS-AHP for solar plant site optimization but did not sufficiently consider spatiotemporal variations pertinent to charging station locations, including future charging demands and traffic patterns [6]. combined GIS, AHP, and TOPSIS for wind farm site suitability analysis, providing a multi-criteria framework but lacking considerations of policies and social acceptance issues in renewable energy deployment.
In transportation infrastructure optimization [30], combined GIS, Artificial Intelligence (AI) and MOORA for bike-share station site selection, sharing similarities with EVCS siting regarding spatial planning, user behavior patterns, and infrastructure coverage. Nonetheless, bike-sharing site selection mainly depends on urban population density and road networks, while EVCS site selection additionally involves electrical infrastructure and renewable energy feasibility. Thus, directly adopting similar methods may be limited. Additionally, these studies rely on historical data and static analyses, whereas optimal EVCS locations require real-time traffic, electrical load variations, and renewable energy supply fluctuations, indicating the necessity for improved dynamic modeling.
To enhance existing methods and improve accuracy, hybrid decision-making models have emerged. For instance [3], combined GIS-MCDM with fuzzy decision-making methods for photovoltaic charging station optimization based on hypothetical data, lacking empirical validation [12]. applied fuzzy VIKOR to island-based EVCS site selection without accounting for evolving traffic demand. By comparison [14], introduced a two-stage multi-objective framework integrating fast-charging stations, photovoltaics, and energy storage systems, simultaneously optimizing charging infrastructure and renewable energy utilization. Nevertheless, this method requires further empirical validation in diverse urban environments.
In summary, GIS-MCDM application in EVCS site selection is well-developed, successfully applied in fields such as landfills, solar power plants, wind farms, and shared mobility. However, existing research faces challenges, including data staticity, limited dynamic optimization, and insufficient social acceptance studies. Future research should integrate real-time data analytics, machine learning optimization, and multi-level site assessments to enhance adaptability and feasibility. Moreover, integration with policy and social factors requires further advancement to ensure equitable, economically viable, and sustainable EVCS networks.
Methodology
To ensure methodological consistency and reproducibility, the Web of Science (WoS) Core Collection was used as the sole data source. The Advanced Search settings were reported verbatim to enable replication. The selected indexes included Science Citation Index Expanded (SCI-EXPANDED), Social Sciences Citation Index (SSCI), Emerging Sources Citation Index (ESCI), and Arts & Humanities Citation Index (A&HCI). The Boolean expression used was (#1 OR #2), where #1: TS=(GIS AND MCDM) and #2 = TS=(electric vehicle charging station sites). Filters were applied for Language = English, Timespan = 2016–2025, and Document Types = Article OR Review. The interface-equivalent form is ALL=((GIS AND MCDM) OR (electric vehicle charging station sites)) AND LA=(English) AND DT=(Article OR Review) AND PY=(2016–2025), with the index restriction applied via the checkboxes listed above. Scope is WoS-only and English-only; formal inference is confined to 2016–2024 with the partial 2025 presented descriptively; the analysis is cross-sectional based on VOSviewer outputs. Comparable EV–sustainability bibliometric studies relying on WoS and a VOSviewer/Bibliometrix workflow have established this as a transparent and reproducible pipeline [33]. As a design choice, the partial year of 2025 is used for descriptive context only and is excluded from trend inference. This decision reduces bias from incomplete annual coverage and enhances reproducibility. The search was executed on 23 February 2025; therefore, the 2025 record represents a partial-year snapshot. To avoid within-year accumulation bias, all formal trend interpretation is based on the 2016–2024 complete interval, and the 2025 value is presented descriptively without being used for statistical inference. Conference and book indexes (CPCI-S, CPCI-SSH, BKCI-S, BKCI-SSH) were not selected, and records unrelated to EVCS applications were removed. This WoS-only design follows established practice in VOS/bibliometric studies [15, 16, 17–18], while the coverage limits of a single-database approach are explicitly acknowledged to enhance transparency and reproducibility. Ultimately, 1336 relevant articles were identified, forming the dataset for research trend analysis.
During data processing, tools such as VOSviewer and Excel were used for data cleaning, organization, and visual analysis. Initially, complete records and citation information from WoS publications, including authors, titles, keywords, abstracts, publication years, and journals, were exported, formatted consistently, and cleansed by removing duplicates and irrelevant articles. Subsequently, keyword co-occurrence analysis using VOSviewer identified core research themes and trends; author collaboration network analysis highlighted key researchers and academic cooperation relationships; national/institutional analysis explored academic contributions and cooperative patterns of various countries and institutions; and citation analysis identified highly cited literature and its impact. Furthermore, Excel was utilized for annual publication statistics, visually presenting the evolving trends of GIS-MCDM research on EVCS site selection. When interpreting the maps, structural features are read directly from the VOSviewer outputs—namely the visible pattern of co-occurrence connections and the color-coded clusters—so that explanations remain consistent with the mapping procedure, without introducing measures outside the VOSviewer outputs. The analysis concentrates on cross-sectional co-occurrence mapping and cluster-based interpretation; temporal evolution overlays and burst-keyword detection are beyond the scope of this paper.
The data collection and analysis methods employed demonstrate scientific rigor and rationality. First, the high-quality WoS database ensures authoritative data derived from peer-reviewed publications. Second, VOSviewer’s network analysis clearly depicts academic collaboration patterns and thematic clusters, enhancing trend visualization clarity. Additionally, combining bibliometric and trend analyses provides multidimensional insights into GIS-MCDM developments and emerging directions. However, the study exhibits certain limitations, including potential temporal lag in WoS data, possible exclusion of recent research, language limitations restricting analyses to English publications, and limited database coverage excluding sources such as Google Scholar or Scopus. Given these boundaries, we interpret the results conservatively and cite WoS-only methodological precedents [15, 16, 17–18] as supporting references. Because the 2025 records are incomplete as of 23 February 2025, they are excluded from formal inference and reported descriptively. As a methodological boundary, the present study does not implement thematic evolution or burst-keyword analysis; future work will integrate these dynamic techniques to trace topic trajectories and emerging fronts. To retain a focused narrative, we provide a concise cross-study alignment in the Discussion and defer a full meta-comparison across EVCS and broader infrastructure bibliometrics to future work.
In conclusion, the systematic bibliometric method applied ensures comprehensive and accurate data, offering a reliable basis for analyzing GIS-MCDM research trends in EVCS site selection and providing significant references for future studies.
Results
[See PDF for image]
Fig. 1
Annual publications (2016–2025). Counts are from WoS; formal trend inference is based on 2016–2024 while the partial 2025 snapshot
This study analyzed the number of publications related to GIS-MCDM in EVCS site selection from 2016 to 2025 based on the collected data (Figure 1). According to the trends, scholarly interest in this research area has exhibited notable changes.
From the data, the publication volume on GIS-MCDM in EVCS site selection has shown a continuous growth trend from 2016 to 2025, peaking notably in 2022 (18.039%) and 2024 (18.563%). Recently, research interest has significantly increased, particularly accelerating from 2019 (6.961%), with publication proportions consistently exceeding 10% between 2021 and 2024. This indicates rising attention to GIS-MCDM applications in EVCS site selection.
The publication volume experienced a slight decline in 2023 (13.249%) but remained at a high level overall, whereas the relatively lower publication rate in 2025 (3.892%) could result from incomplete data coverage at present. Consistent with the Methods, formal trend statistics are reported for 2016–2024, whereas the 2025 partial-year value is reported descriptively and excluded from formal inference. This reporting choice aligns with the methodological design and prevents over-interpretation of an incomplete year. Overall, GIS-MCDM for EVCS site selection has entered a phase of rapid growth and has emerged as a prominent research topic within sustainable energy, intelligent transportation, and spatial optimization.
Regarding research themes, the increasing application of GIS-MCDM methods may be driven by supportive renewable energy policies, increasing demand for charging infrastructure construction, and the growing sophistication of intelligent decision-making in site selection. Particularly, the growth in 2022 and 2024 might be closely linked to global renewable energy deployment, EV market expansion, and smart city development initiatives. Future studies in this domain are expected to further integrate AI, spatial big data, and dynamic charging station optimization to deepen GIS-MCDM application in renewable energy infrastructure development.
The peak research period (2021–2024) can be attributed to multiple factors. Firstly, rapid expansion in the EV market has increased governmental and industry demands for optimized charging station layouts, stimulating relevant academic research. Secondly, diversified data acquisition methods, the integration of high-resolution remote sensing, geographic big data, and machine learning technologies have expanded the applicability of GIS-MCDM decision models in EVCS site selection. Additionally, supportive government policies and sustainable urban development strategies have contributed to increased research activity.
Notably, the slight decrease in 2023 research publications may relate to market dynamics, policy adjustments, or periodic shifts in research focus. Nevertheless, the rebound in 2024 indicates ongoing research significance and development potential of GIS-MCDM applications in EVCS site selection. With advancements in smart cities, autonomous electric vehicles, and smart grid technologies, this research area is expected to maintain high academic attention in the forthcoming years.
[See PDF for image]
Fig. 2
Author co-authorship network. (node size = output; links = collaborations; colors = clusters)
Figure 2 visualizes the co-authorship structure among prolific authors within GIS-MCDM research in EVCS site selection. As depicted, Pradhan B published the most related articles (13 publications), reflecting significant scholarly influence in the domain. Following this, Wu YN ranks second with 10 articles, while Deveci M and Xu CB each contributed 9 articles, establishing themselves as important contributors. Other scholars, including Arabameri A, Pourghasemi HR, Kabak M, Wang CN, Sánchez-Lozano JM, and Özceylan E, have each published 8 articles, demonstrating notable but not dominant contributions. This balanced distribution suggests the domain is driven collectively rather than dominated by a few individuals. At the author level, we further qualify prolific contributors using WoS ‘Times Cited’ and average citations per item (ACI): for example, Pradhan B and Wu YN not only publish extensively but also accumulate substantial citations with stable ACI values, indicating sustained influence in this topic.
The distribution further indicates a strong collaborative trend among multiple scholars, highlighting the broad applicability of GIS-MCDM for EVCS siting. High-output researchers such as Pradhan B and Wu YN potentially played pivotal roles in refining spatial analyses, enhancing MCDM models, and evaluating infrastructure suitability, guiding research toward greater sophistication.
Given the steady research activity and balanced academic contribution, GIS-MCDM for EVCS site selection has long-term development potential. Advances in GIS technology, the accumulation of spatial data, and refined MCDM methods suggest ongoing interest. Future research directions might increasingly incorporate AI, traffic big data, and smart grid optimization technologies, improving scientific rigor and decision efficiency for EVCS site selection.
[See PDF for image]
Fig. 3
Country collaboration network. (node size = output; links = collaborations; colors = clusters)
From the network generated using VOSviewer (Figure 3), collaborative relationships among different countries in GIS-MCDM research for EVCS site selection are observed. Node sizes represent the quantity of publications, link density indicates collaboration intensity, and colors reflect distinct research collaboration clusters.
As illustrated, China and the United States emerge as countries with the highest research outputs, occupying central roles in international collaboration networks and indicating their dominant positions in GIS-MCDM applications for EVCS site selection. India, Iran, and Australia also make considerable contributions, forming relatively independent collaboration networks. Additionally, Spain and the United Kingdom constitute active collaboration centers, exhibiting strong intra-European academic linkages. The network indicates dense regional ties but comparatively limited cross-regional links, implying that broader collaboration would support the transferability and external generalization of EVCS site-selection findings.
The topology of this collaboration network demonstrates significant internationalization and multicentric characteristics. Extensive collaborative clusters form between multiple nations, exemplified by China–U.S. collaborations extending across Asia, Europe, and America. Additionally, Iran, India, and Saudi Arabia have established stable regional collaboration networks, possibly related to their national policies on renewable energy infrastructure development.
The collaborative network reveals the widespread application of GIS-MCDM globally, especially in countries rapidly advancing renewable energy initiatives, including China, the United States, India, and Australia. This correlates with national policies on electric vehicle penetration, charging network planning, and smart city development. Overall, GIS-MCDM research for EVCS site selection demonstrates robust international cooperation, with academic exchanges among different nations driving ongoing progress.
[See PDF for image]
Fig. 4
Top publishing countries (share of WoS records, 2016–2024)
According to data from the Web of Science (Figure 4), China (18.713%), India (16.692%), and Iran (13.323%) have significantly contributed to GIS-MCDM research for EVCS site selection, publishing 250, 223, and 178 studies, respectively, reflecting active research engagement. This is likely related to renewable energy policies, EV adoption rates, and the maturity of GIS technologies. The United States (9.955%) also shows substantial research output, indicating sustained investment in intelligent transport infrastructure and spatial optimization.
At the institutional level, WoS affiliation data indicate that outputs are concentrated in several comprehensive universities and engineering schools, with relatively small gaps among the most productive units and no single dominant institution. To maintain concision and avoid redundant listings, we summarize the overall pattern in text, using publication volume and citation performance as impact references.
By comparison, Saudi Arabia (5.090%), Turkey (6.213%), Australia (4.042%), the United Kingdom (3.817%), and Spain (3.668%) have fewer publications yet contribute meaningfully. European countries (Spain, UK) and Middle Eastern nations (Saudi Arabia, Iran) show lower but growing research activities driven by renewable energy policy and EV market development.
Thematically, countries with greater research outputs typically have higher EV adoption rates, mature GIS applications, and infrastructure development demands. Active research in China, India, and the US aligns closely with accelerating urbanization, increased charging network optimization needs, and supportive government policies. Research in Iran and Saudi Arabia likely reflects policy shifts and the application of MCDM in energy management. GIS-MCDM application globally is expected to expand further, integrating intelligent transportation systems and renewable energy infrastructure optimization.
[See PDF for image]
Fig. 5
Journal distribution by number of publications (top venues)
From bibliometric data (Figure 5), the research involving GIS-MCDM in EVCS site selection spans numerous journals, showing considerable variation in publication quantity. Energies has the highest number of publications, with 58 articles (4.341%), demonstrating substantial influence in energy management and GIS applications. Other notable journals include Sustainability (51 articles, 3.817%), Renewable Energy (36 articles, 2.695%), and Energy (36 articles, 2.695%), reflecting the extensive application of GIS-MCDM in sustainable energy and renewable energy optimization.
Journals such as Applied Energy (31 articles, 2.320%), Environmental Science and Pollution Research (34 articles, 2.545%), Journal of Cleaner Production (34 articles, 2.545%), and Sustainable Cities and Society (31 articles, 2.320%) contribute fewer publications but cover critical topics like environmental pollution, clean production, and urban sustainability. Notably, IEEE Access (22 articles, 1.647%) and Natural Hazards (18 articles, 1.347%) provide perspectives on technology and risk analysis in GIS-MCDM applications for EVCS.
Thematically, GIS-MCDM research for EVCS site selection primarily focuses on energy management, sustainable development, and environmental optimization. Journals with higher publication rates are involved in renewable energy utilization, intelligent transportation, and sustainable urban development, suggesting the method’s broad applicability beyond EVCS optimization to energy distribution, environmental impact assessments, and clean production. Future studies might further extend into intelligent grids, carbon-neutral policies, and spatial big data analytics, enhancing the scientific rigor and decision efficiency of GIS-MCDM in renewable energy infrastructure planning.
[See PDF for image]
Fig. 6
Keyword co-occurrence network (VOSviewer; colors = clusters)
Figure 6 visualizes the core keyword co-occurrence network in GIS-MCDM research for EVCS site selection, generated using VOSviewer. The clusters, distinguished by color, represent thematic keyword groupings, with node sizes indicating keyword frequency and line thickness representing association strength among keywords. Analysis indicates that research in this domain primarily centers around four key thematic clusters.
Structurally, the map exhibits cohesive links within clusters and comparatively sparse ties across clusters. Reading the maps at the cluster level—with compact intra-cluster ties and limited yet meaningful inter-cluster connections—helps trace the four thematic strands while avoiding over-generalization from global coefficients. A limited set of terms provides visible cross-cluster connections, delineating the pathways among the four themes. These structural patterns underpin the subsequent cluster-wise narration as coherent topic blocks with a small number of meaningful inter-cluster links.
(Red cluster) Layout and planning of electric vehicle charging stations:
This cluster includes keywords such as “charging stations,” “electric vehicles,” “planning,” and “energy management,” demonstrating that layout optimization is a significant GIS-MCDM application, involving aspects of energy management, charging infrastructure optimization, and traffic flow analysis.
(Blue cluster) GIS and Multi-Criteria Decision-Making (MCDM):
This cluster, characterized by keywords such as “GIS”, “AHP”, “TOPSIS”, and “remote sensing” indicates the widespread use of GIS combined with MCDM methods. The integration of methods like AHP, TOPSIS, Stepwise Weight Assessment Ratio Analysis (SWARA) and Best-Worst Method (BWM) enhances the scientific rigor of site selection processes.
(Green cluster) Spatial analysis and sustainable development:
Keywords such as “site selection,” “decision-making,” “sustainability,” “photovoltaic,” and “wind farms” indicate that GIS-MCDM methodologies are extensively applied to renewable energy facility site selection. This cluster overlaps significantly with EVCS research, underscoring the interdisciplinary nature of renewable energy integration.
(Yellow cluster) Uncertainty and environmental factor assessment:
Including keywords like “uncertainty,” “risk assessment,” “spatial analysis,” and “fuzzy MCDM,” this cluster highlights research attention toward addressing environmental uncertainties, risk evaluation, and spatial analysis, enhancing reliability and sustainability in site selection decisions.
Relationship between data analysis and research themes
GIS-MCDM research in EVCS site selection primarily revolves around four directions: charging infrastructure optimization, multi-criteria decision methods, renewable energy integration, and uncertainty analysis. This indicates that GIS-MCDM is not confined to static site optimization but is broadly applicable to dynamic charging demand prediction, renewable energy integration, and environmental sustainability assessments.
The cross-disciplinary nature of research indicates future GIS-MCDM applications may integrate intelligent optimization algorithms, real-time traffic data, and spatial big data analytics to enhance the intelligence of EVCS networks. Concurrently, comprehensive consideration of environmental and social factors will drive GIS-MCDM methodologies toward more holistic and intelligent decision-making systems, responding effectively to evolving renewable transportation infrastructure demands.
[See PDF for image]
Fig. 7
Keyword hotspot map (VOSviewer heatmap overlay)
Figure 7 illustrates the core research hotspots in GIS-MCDM for electric vehicle charging station site selection using a heatmap visualization generated by VOSviewer. Regions with brighter colors indicate higher research interest, whereas darker regions reflect lower engagement. Overall, the primary research hotspots include GIS, MCDM, optimization of charging station locations, integration of renewable energy, and environmental impact analyses, reflecting diversified research directions in this field.
The core of research hotspots is concentrated on the integration of GIS with MCDM methods, with prominent keywords such as “GIS,” “AHP,” “multicriteria decision analysis,” and “decision-making” indicating the essential roles these technologies and methods play in optimizing charging station site selection. Decision-making methods such as AHP, Weighted Linear Combination (WLC), and SWARA are particularly prevalent, underscoring their importance in enhancing the scientific validity and rationality of site selection decisions.
In addition to methodological research, station layout optimization represents a critical research direction. Keywords such as “electric vehicles,” “charging stations,” “optimization,” and “planning” demonstrate researchers’ focus on issues like rational layout, traffic flow analysis, charging demand forecasting, and energy allocation, aiming to enhance infrastructure efficiency and coverage. The hotspots converge on GIS–MCDM integration, station layout optimization, and energy-related coupling, providing a coherent agenda that is consistent with the synthesis reported in this study.
Concurrent with site selection, research also investigates renewable energy integration and environmental assessments. Keywords such as “solar,” “wind farm,” “biomass,” “suitability analysis,” and “risk assessment” indicate that GIS-MCDM methodologies are extensively utilized for siting renewable energy facilities, environmental sustainability evaluations, and land-use optimization. These contribute synergistically to the coordinated development of electric vehicle charging infrastructure and renewable energy resources.
Additionally, uncertainty and risk management have garnered scholarly attention, as shown by keywords like “uncertainty,” “logistic regression,” “sensitivity analysis,” and “risk assessment.” Addressing uncertainties related to environmental, social, and technological factors through methods like fuzzy MCDM and probabilistic analysis is essential for enhancing the robustness of site selection decisions.
Overall, GIS-MCDM research on EVCS site selection has formed a recognizable framework, covering spatial data processing, station optimization, renewable energy integration, environmental impact assessments, and uncertainty analysis. Future research may integrate AI, real-time traffic data, and large-scale spatial data analytics to elevate the sophistication and efficiency of site selection methodologies. Simultaneously, comprehensive consideration of social, economic, and environmental factors will further promote GIS-MCDM approaches toward more multi-dimensional and intelligent decision-making systems, meeting the growing demands of electric vehicle charging infrastructure.
[See PDF for image]
Fig. 8
Document co-citation network. (node size = citations; colors = clusters)
Figure 8 presents a document co-citation network for GIS-MCDM research in EVCS site selection, generated using VOSviewer. At the document level, based on the WoS ‘Times Cited’ indicator, the representative top-cited studies include [34] IEEE Transactions on Power Delivery (454 citations) [6], Applied Energy (307 citations) [29], Applied Energy (225 citations), and [35] IEEE Transactions on Power Delivery (208 citations). Collectively, these works have shaped the methodological and applied foundations of GIS-MCDM for EVCS siting. Distinct clusters (color-coded) represent different research directions or academic subfields. Node sizes indicate citation frequency, while connecting lines illustrate co-citation relationships, reflecting how frequently documents are cited simultaneously in the literature.
Overall, the co-cited documents can be categorized into three main academic clusters:
(Red cluster) Theoretical foundations of MCDM methodologies
This cluster revolves around foundational theories of MCDM methods, especially [11], AHP, and includes related mathematical modeling, optimization approaches, and decision-support techniques. These foundational works have been extensively applied to spatial optimization, including EVCS site selection.
(Green cluster) renewable energy and EVCS siting applications
This cluster includes practical applications of GIS-MCDM in renewable energy siting, such as studies by [32] and [36] on solar and wind energy site selection, demonstrating a strong connection to EVCS research. GIS-MCDM methods thus serve not only spatial optimization for EV infrastructure but also broader renewable energy site selection objectives supporting sustainable transportation. This pattern is consistent with a 2015–2025 review indicating that AHP, TOPSIS, WASPAS, and MOORA remain central in sustainable-energy decision-making, with increasing integration of fuzzy designs—an arc that parallels the EVCS trajectory mapped in this study [37].
(Blue cluster) Empirical studies and application optimization
Focused on empirical research and practical optimization applications, represented by studies such as [6] and [30], this cluster addresses GIS-MCDM for real-world EVCS site selection and energy management optimization, including traffic flow analysis, land use assessment, and power network optimization, reflecting the trend toward enhanced decision accuracy and efficiency.
Relationship of research trends and themes
The co-citation analysis indicates GIS-MCDM research in EVCS site selection has formed a structured framework consisting of theoretical foundations, applied studies, and empirical optimization. Research advances have relied significantly on mathematical modeling theories underpinning MCDM, alongside expansions into renewable energy siting, charging infrastructure layout optimization, and data-driven enhancements. Future GIS-MCDM research may further integrate intelligent transportation data, real-time optimization algorithms, and renewable energy integration, improving charging infrastructure sustainability and intelligence.
[See PDF for image]
Fig. 9
Author co-authorship network (overlay by publication year, 2020–2025)
Figure 9 visualizes the author collaboration network within the GIS-MCDM research domain for EVCS site selection using VOSviewer. Nodes represent researchers, connections indicate collaborative relationships, and colors reflect publication years (2020–2025), with greener or yellower hues indicating more recent studies.
The network indicates dispersed collaborative relationships, with only a few closely connected author clusters. Pradhan Biswajeet occupies a central role, maintaining strong collaborations with Pourghasemi Hamid Reza, Arabameri Alireza, and Blaschke Thomas, reflecting significant academic influence in the GIS-MCDM research domain. Blaschke Thomas and Arabameri Alireza also substantially contribute to GIS methodologies, MCDM model optimization, and spatial planning for EV infrastructure.
In contrast, Mokarram Marzieh appears as the sole researcher in a newer research period (2024–2025), implying a focus on recent developments such as AI-driven GIS-MCDM and intelligent charging network planning. Mokarram’s limited collaboration indicates an independent research trajectory, suggesting potential for future collaboration growth. Strengthening international collaboration would improve comparability of results and support cumulative progress in EVCS site-selection research.
Overall, author networks indicate GIS-MCDM research for EVCS siting is currently led by key researchers rather than large collaborative groups. Future research may increasingly adopt interdisciplinary, multi-team collaborations, integrating AI, big data analytics, and real-time traffic data to enhance scientific rigor and practical intelligence in EVCS infrastructure planning.
Discussion
Principal findings
The integration of GIS and MCDM has become a widely used approach for selecting EVCS locations. Methodologically, EVCS-focused fuzzy MCDM work also identifies technology improvement and charging-station expansion as high-impact levers, reinforcing our cluster-wise reading of the maps [10]. Recent reviews and bibliometric studies report patterns consistent with our results. These studies also adopt VOSviewer-based co-occurrence mapping, aligning methodologically with our pipeline. Distinctly, the present work couples GIS with MCDM and narrows the lens to EVCS site selection, extends the window to 2016–2025, and provides a cluster-wise structural reading of the maps rather than a broad meta-benchmark across domains [15, 16, 17–18].
Theoretical gaps remain. First, dynamic uncertainty—time-varying demand, grid loading, and renewable intermittency—has not been routinely embedded in decision formulations, with many designs still anchored in static inputs [3, 21, 23]. Second, distributive and procedural equity objectives are rarely formalised as constraints or outcome metrics despite their practical salience [27]. Third, external validity and replicability are limited by hypothetical datasets and sparse cross-city validation [14]. Finally, comparative robustness across weighting/ranking families (e.g., AHP/TOPSIS vs. CRITIC/Entropy vs. fuzzy variants) is not consistently reported, motivating transparent sensitivity analyses alongside open data and query disclosures [15, 16–17].
In this context, AI-driven site optimization can be considered as a complementary layer to the current GIS + MCDM workflow, supporting periodic updates, sensitivity checks, and re-prioritization when new evidence becomes available. Recent evidence on machine-learning approaches for EVCS siting further motivates this prospective linkage [38]. A systematic assessment of this interface is identified as future work and falls outside the remit of the present bibliometric analysis.
For urban planners, the mainstream GIS + MCDM workflow identified in this study can be adopted as a standard, auditable pipeline for EVCS site screening and prioritization, with explicit recording of data sources and parameter settings to ensure transparency and reproducibility. For policymakers, the growth and collaboration patterns encourages cross-department and university–industry collaboration and requires public reporting of methodological choices (database platform, query terms, and time window) when commissioning EVCS siting studies. For industry stakeholders, aligning feasibility studies with the GIS + MCDM approach and partnering with leading research teams can reduce assessment costs and accelerate deployment. Periodic updates of analyses are also recommended.
Its applications have expanded to multi-stakeholder evaluation, intelligent optimization, and renewable energy integration. However, international collaboration is limited, primarily occurring at regional levels, with little global data sharing. Evidence from ASEAN emerging economies attributes adoption gaps to high upfront costs, sparse charging networks, and fragmented policy, which helps contextualize the regional skew and constrained cross-border ties observed here [36]. Future research should enhance international cooperation and establish global data-sharing platforms to improve site selection model adaptability and applicability.
Research hotspots in GIS-MCDM
Methodological applications
GIS-MCDM integrates spatial analysis with multi-criteria decision-making to optimize EVCS placement. It has become a standard tool in this field. Common methods include AHP, TOPSIS, SWARA, WLC, and BWM. These techniques ensure scientifically sound and rational site selection.
Multi-factor considerations
EVCS site selection considers multiple factors. Geographic criteria include road networks, grid access, and land availability. Socioeconomic factors encompass population density, traffic flow, land use, and electricity demand. Environmental considerations such as air quality and ecological sensitivity are also essential. A multi-dimensional decision-making approach enhances precision and sustainability in site planning.
Emerging directions
Recent research has identified several emerging directions, reflecting the dynamic nature of EVCS planning.
Renewable energy integration
Recent research focuses on incorporating renewable energy sources into EVCS site selection. Solar- and wind-powered charging stations aim to reduce reliance on conventional power grids and improve sustainability. This approach supports green transportation and contributes to a low-carbon energy transition. This aligns with global sustainability goals, as evidenced by keywords like “solar farm” and “wind farm” in recent studies.
Transition from static to dynamic optimization
With rising EV adoption, charging demand has become increasingly uncertain. Traditional static site selection based on historical data is shifting towards dynamic optimization. Recent studies incorporate real-time traffic flow, grid load, and weather conditions. Leveraging big data and real-time analytics minimizes site selection errors, enhances flexibility, and optimizes urban grid management. This shift is crucial for accommodating fluctuations in EV adoption and charging demand, ensuring adaptive infrastructure planning.
Intelligent systems and visualization
Machine learning, deep learning, and AI are being integrated into GIS-MCDM models to improve decision-making accuracy. Additionally, Virtual Reality (VR) and Augmented Reality (AR) applications are also under exploration for simulating charging station layouts and enhancing user experience. However, VR/AR remains underutilized in this field. Future research should further investigate their potential in site selection.
Interdisciplinary integration
GIS-MCDM research increasingly spans multiple disciplines, including energy management, urban planning, and environmental science. This interdisciplinary approach enhances site selection models and promotes collaborative infrastructure development, integrating diverse perspectives to address multifaceted challenges.
Addressing uncertainty
Given the uncertainty in future EV market developments, research must employ methods such as fuzzy mathematics, logistic regression, and probabilistic analysis to handle uncertain environmental variables. These techniques are crucial for mitigating the effects of uncertainty, enhancing the resilience of site selection decisions over the long term, and ensuring adaptability to changing conditions.
Research trends
From 2016 to 2025, GIS-MCDM research on EVCS site selection has evolved from basic methodological applications to comprehensive optimization. Early studies focused on implementing MCDM techniques (e.g., AHP, TOPSIS) in GIS environments. More recent work has expanded into intelligent optimization, renewable energy integration, and dynamic site selection. Topics such as photovoltaic charging, backup power systems, and traffic flow prediction are gaining attention. Additionally, socioeconomic factors, including policy and urban planning, play an increasingly significant role in decision-making.
Future research on GIS-MCDM for EVCS site selection is expected to evolve along three key directions. First, intelligent optimization will become increasingly prominent, with the integration of machine learning algorithms and Internet of Things (IoT) technologies. These tools will enable real-time responsiveness to dynamic data such as traffic flow, electricity demand, and user charging behaviors, ultimately allowing for more adaptive and efficient site selection strategies.
Second, there is likely to be a stronger emphasis on sustainability and equity. As environmental and social considerations become central to urban planning, future GIS-MCDM frameworks will incorporate criteria related to carbon emissions, renewable energy integration, and equitable access to infrastructure. This ensures that EVCS development aligns not only with environmental goals but also with broader efforts to reduce spatial and social disparities.
Lastly, participatory decision-making is anticipated to play a more significant role. Future research will explore methods to systematically include the perspectives of stakeholders such as city planners, energy providers, and local communities. This trend reflects a growing recognition of the importance of transparency and inclusiveness in infrastructure planning. Emerging tools and platforms will likely support more interactive and collaborative GIS environments, enabling broader engagement in the decision-making process. Taken together, the results provide an EVCS-oriented GIS-MCDM workflow, a reproducible reporting standard for commissioned analyses, and collaboration signals from author- and country-level networks that can inform near-term planning and policy allocation.
Conclusion
This study conducted a bibliometric analysis of 1336 publications from 2016 to 2025 in the Web of Science database and visualized research trends using VOSviewer. Drawing on the evidence, three patterns underpin our concluding insights. First, the country collaboration and output shares indicate regional clustering, with leading contributions concentrated in a small set of countries (e.g., China 18.713%), and sparse cross-regional ties in the network. Second, the journal distribution is anchored in energy-sustainability venues (e.g., Applied Energy, 31 items), linking methodological advances to deployment contexts. Third, the keyword co-occurrence map consolidates four cohesive clusters—EVCS layout/planning, GIS–MCDM methods, spatial sustainability, and uncertainty/risk—indicating a maturing field that couples methodological refinement with implementation pathways. The findings highlight the increasing adoption of GIS-MCDM as a key approach for EVCS site selection, with applications expanding into multi-criteria decision optimization, intelligent methods, and renewable energy integration.
Future work should be evidence-driven along five lines. (1) Strengthen international collaboration and open datasets to overcome the regional clustering observed in the country network and to improve external generalisability. (2) Embed dynamic uncertainty—temporal demand, grid loading, and renewable intermittency—as explicit model components, consistent with the “uncertainty/risk” keyword cluster. (3) Formalise equity as constraints and outcome metrics (e.g., coverage in disadvantaged areas), aligning site-selection with spatial-justice objectives signalled by country and institutional distributions. (4) Report cross-model robustness and external validity through multi-city or out-of-sample tests, bridging methodological families (AHP/TOPSIS, objective weighting such as CRITIC/entropy, and fuzzy/intuitionistic variants). (5) Integrate energy–transport coupling (PV, storage, and grid constraints) and participatory inputs to reflect the journal landscape’s focus on sustainable energy deployment and to support policy-ready decisions. It should be noted that our dataset is confined to the WoS Core Collection; future work will integrate Scopus and IEEE Xplore through de-duplication and harmonized export to further enhance coverage and representativeness.
Acknowledgements
The authors would like to thank Universiti Sains Malaysia and the Ministry of Higher Education Malaysia for supporting this research under the Fundamental Research Grant Scheme (FRGS) project title “Development of Integrated Geographic Information System and Agent-Based Model for Forecasting A Sustainable Urban Future” [Grant number: FRGS/1/2023/SS07/USM/02/1].
Author contributions
Narimah Samat conceptualized the study and wrote the initial manuscript draft. Wenhao Li and Mohd Amirul Mahamud performed data analysis. Mou Leong Tan contributed to the methodology and visualization. All authors read and approved of the final manuscript.
Funding
This research is supported by Universiti Sains Malaysia and the Ministry of Higher Education Malaysia through the Fundamental Research Grant Scheme (FRGS), project title “Development of Integrated Geographic Information System and Agent-Based Model for Forecasting A Sustainable Urban Future” [Grant number: FRGS/1/2023/SS07/USM/02/1].
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent to publish
Not applicable.
Competing interests
The authors declare no competing interests.
Clinical trial number
Not applicable.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
1. Rashmitha, Y; Sushma, MB; Roy, S. A novel multi-criteria framework for selecting optimal sites for electric vehicle charging stations from a sustainable perspective: evidence from India. Environ Dev Sustain; 2024; [DOI: https://dx.doi.org/10.1007/s10668-024-05746-4]
2. Zhao, H; Gao, J; Cheng, X. Electric vehicle solar charging station siting study based on GIS and multi-criteria decision-making: A case study of China. Sustainability; 2023; 15,
3. Zhou, J; Wu, Y; Wu, C; He, F; Zhang, B; Liu, F. A geographical information system-based multi-criteria decision-making approach for location analysis and evaluation of urban photovoltaic charging station: A case study in Beijing. Energy Conv Manag; 2020; 205, 112340. [DOI: https://dx.doi.org/10.1016/j.enconman.2019.112340]
4. Kaya, O; Alemdar, KD; Atalay, A; Codur, MY; Tortum, A. Electric car sharing stations site selection from the perspective of sustainability: A GIS-based multi-criteria decision making approach. Sustain Energy Technol Assess; 2022; 52, 102026. [DOI: https://dx.doi.org/10.1016/j.seta.2022.102026]
5. Koirala, K; Tamang, M; Shabbiruddin,. Planning and establishment of battery swapping station—A support for faster electric vehicle adoption. J Energy Storage; 2022; 51, 104351. [DOI: https://dx.doi.org/10.1016/j.est.2022.104351]
6. Guo, S; Zhao, H. Optimal site selection of electric vehicle charging station by using fuzzy TOPSIS based on sustainability perspective. Appl Energy; 2015; 158, pp. 390-402. [DOI: https://dx.doi.org/10.1016/j.apenergy.2015.08.082]
7. Liu, H-C; Yang, M; Zhou, M; Tian, G. An integrated multi-criteria decision making approach to location planning of electric vehicle charging stations. IEEE Trans Intell Transp Syst; 2019; 20,
8. Wu, Y; Xie, C; Xu, C; Li, F. A decision framework for electric vehicle charging station site selection for residential communities under an intuitionistic fuzzy environment: A case of Beijing. Energies; 2017; 10,
9. Biswas, A; Gazi, KH; Sankar, PM; Ghosh, A. A decision-making framework for sustainable highway restaurant site selection: AHP-TOPSIS approach based on the fuzzy numbers. Spectr Oper Res; 2025; 2,
10. Eti, S; Dinçer, H; Yüksel, S; Gökalp, Y. A new fuzzy decision-making model for enhancing electric vehicle charging infrastructure. Spectr Decis Mak Appl; 2025; 2,
11. Saaty, TL. The analytic hierarchy process (AHP). J Oper Res Soc; 1980; 41,
12. Dang, R; Li, X; Li, C; Xu, C. A MCDM framework for site selection of Island photovoltaic charging station based on new criteria identification and a hybrid fuzzy approach. Sustainable Cities Soc; 2021; 74, 103230. [DOI: https://dx.doi.org/10.1016/j.scs.2021.103230]
13. Zhang, Y; Teoh, BK; Zhang, L. Integrated bayesian networks with GIS for electric vehicles charging site selection. J Clean Prod; 2022; 344, 131049. [DOI: https://dx.doi.org/10.1016/j.jclepro.2022.131049]
14. Kumar, N; Kumar, T; Nema, S; Thakur, T. A multiobjective planning framework for EV charging stations assisted by solar photovoltaic and battery energy storage system in coupled power and transportation network. Int J Energy Res; 2022; 46,
15. Banegas, J; Mamkhezri, J.
16. Barbosa, W; Prado, T; Batista, C; Câmara, JC; Cerqueira, R; Coelho, R; Guarieiro, L. Electric vehicles: bibliometric analysis of the current state of the Art and perspectives. Energies; 2022; 15,
17. Kut, P; Pietrucha-Urbanik, K. Bibliometric analysis of multi-criteria decision-making (MCDM) methods in environmental and energy engineering using CiteSpace software: identification of key research trends and patterns of international Cooperation. Energies; 2024; 17,
18. Şimşek, Aİ; Taşdemir, Desticioğlu; Koç, E.
19. Hosseini, S; Sarder, MD. Development of a bayesian network model for optimal site selection of electric vehicle charging station. Int J Electr Power Energy Syst; 2019; 105, pp. 110-22. [DOI: https://dx.doi.org/10.1016/j.ijepes.2018.08.011]
20. Sadeghi-Barzani, P; Rajabi-Ghahnavieh, A; Kazemi-Karegar, H. Optimal fast charging station placing and sizing. Appl Energy; 2014; 125, pp. 289-99. [DOI: https://dx.doi.org/10.1016/j.apenergy.2014.03.077]
21. Tu, W; Li, Q; Fang, Z; Shaw, S; Zhou, B; Chang, X. Optimizing the locations of electric taxi charging stations: A spatial-temporal demand coverage approach. Transp Res Part C: Emerg Technol; 2016; 65, pp. 172-89. [DOI: https://dx.doi.org/10.1016/j.trc.2015.10.004]
22. Tulpule, PJ; Marano, V; Yurkovich, S; Rizzoni, G. Economic and environmental impacts of a PV powered workplace parking garage charging station. Appl Energy; 2013; 108, pp. 323-32. [DOI: https://dx.doi.org/10.1016/j.apenergy.2013.02.068]
23. Shepero, M; Munkhammar, J; Widén, J; Bishop, JDK; Boström, T. Modeling of photovoltaic power generation and electric vehicles charging on city-scale: A review. Renew Sustain Energy Rev; 2018; 89, pp. 61-71. [DOI: https://dx.doi.org/10.1016/j.rser.2018.02.034]
24. Farah, RN; Syahirah, NA; Misron, N; Azmi, MSM; Karim, NSA; Husin, NM. A need analysis of the criteria involved in determining suitable locations for photovoltaic electric vehicle charging stations in Malaysia. Malaysian J Fundamental Appl Sci; 2023; 19,
25. Ghosh, A; Ghorui, N; Mondal, SP; Kumari, S; Mondal, BK; Das, A; Gupta, MS. Application of hexagonal fuzzy MCDM methodology for site selection of electric vehicle charging station. Mathematics; 2021; 9,
26. Guler D, Yomralioglu T. (2018). Optimal location selection for electric vehicle charging stations using GIS. In: T. Bandrova & M. Konecny, editors, 7th International Conference on Cartography and GIS, 1(2):340–345. Bulgarian Cartographic Assoc. https://www.webofscience.com/wos/woscc/full-record/WOS:000526176700037
27. Mahdy, M; Bahaj, AS; Turner, P; Wise, N; Alghamdi, AS; Hamwi, H. Multi-criteria decision analysis to optimise siting of electric vehicle charging points—Case study Winchester district. UK Energies; 2022; 15,
28. Chang, N-B; Parvathinathan, G; Breeden, JB. Combining GIS with fuzzy multicriteria decision-making for landfill siting in a fast-growing urban region. J Environ Manage; 2008; 87,
29. Al Garni, HZ; Awasthi, A. Solar PV power plant site selection using a GIS-AHP based approach with application in Saudi Arabia. Appl Energy; 2017; 206, pp. 1225-40. [DOI: https://dx.doi.org/10.1016/j.apenergy.2017.10.024]
30. Kabak, M; Erbaş, M; Çetinkaya, C; Özceylan, E. A GIS-based MCDM approach for the evaluation of bike-share stations. J Clean Prod; 2018; 201, pp. 49-60. [DOI: https://dx.doi.org/10.1016/j.jclepro.2018.08.033]
31. Goswami, SS; Tapankumar, T; Naik, NCK; Gowrishankar, J; Bhosle, N; Singh, A; Raju, GS; Nagesh, D; Santhosh, AJ. Multi-model MCDM framework for sustainable renewable energy selection in india: integrating CRITIC-EDAS-CODAS-CoCoSo. Discover Sustain; 2025; 6,
32. Sanchez-Lozano, JM; Teruel-Solano, J; Soto-Elvira, PL; Socorro Garcia-Cascales, M. Geographical information systems (GIS) and multi-criteria decision making (MCDM) methods for the evaluation of solar farms locations: case study in south-eastern Spain. Renew Sustain Energy Rev; 2013; 24, pp. 544-56. [DOI: https://dx.doi.org/10.1016/j.rser.2013.03.019]
33. Shunmathy, V; Selvam, V. A bibliometric review exploring the nexus between environmental sustainability and electric four-wheeler. Discover Sustain; 2025; 6,
34. Liu, Z; Wen, F; Ledwich, G. Optimal planning of electric-vehicle charging stations in distribution systems. IEEE Trans Power Delivery; 2013; 28,
35. Etezadi-Amoli, M; Choma, K; Stefani, J. Rapid-charge electric-vehicle stations. IEEE Trans Power Delivery; 2010; 25, pp. 1883-7. [DOI: https://dx.doi.org/10.1109/TPWRD.2010.2047874]
36. Lin, OZ; Stepanec, L; Koutroulis, E; Juchelkova, D; Aye, HY; de Luna Era, M; Kheang, PV. Barriers to electric vehicle adoption in ASEAN emerging economies: comparative analysis of Cambodia, Myanmar, and the Philippines. Discover Sustain; 2025; 6,
37. Sahoo, SK; Choudhury, BB; Dhal, PR; Hanspal, MS. A comprehensive review of multi-criteria decision-making (MCDM) toward sustainable renewable energy development. Spectr Oper Res; 2025; 2,
38. Huang, Y; Ren, H; Jia, X; Yu, X; Xie, D; Zou, Y; Chen, D; Yang, Y. A spatially aware machine learning method for locating electric vehicle charging stations. World Electr Veh J; 2025; 16,
© The Author(s) 2025. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.