Geographic information system–based multi-criteria decision framework for siting photovoltaic-powered electric vehicle charging stations

• This study presents a systematic literature review (SLR) of 43 peer-reviewed articles (2010–2024) on PV-powered electric vehicle charging station (EVCS) siting using GIS and MCDM methods. • The most frequently used methods are AHP and TOPSIS, while GIS supports spatial visualization and suitability mapping. • Emerging frameworks increasingly incorporate AI and machine learning to enhance predictive accuracy and adaptability. • Sustainability dimensions (environmental, social, and economic) are integrated unevenly, with social equity and real-time data often underrepresented. • The study proposes a hybrid GIS–MCDM–AI framework as a future direction for context-sensitive, data-driven, and sustainable EVCS planning. The accelerating adoption of electric vehicles (EVs) has created an urgent demand for efficient and sustainable charging infrastructure. Among the viable alternatives, photovoltaic (PV)-powered electric vehicle charging stations (EVCS) stand out for their alignment with global decarbonization and renewable energy targets. However, determining optimal EVCS locations requires integrating diverse criteria, including spatial suitability, technical feasibility, environmental impact, socio-economic considerations, and grid connectivity. This study presents a systematic literature review (SLR) of 43 peer-reviewed articles published between 2010 and 2024, applying the PRISMA methodology to examine the use of geographic information systems (GIS) and multi-criteria decision-making (MCDM) approaches in PV-based EVCS site selection. The analysis is structured around five key themes: analytical frameworks, dominant MCDM methods, artificial intelligence (AI) integration, spatial data utilization, and methodological challenges. Results indicate that AHP and TOPSIS are the most widely applied MCDM techniques, enabling structured evaluation of multiple criteria, while GIS is extensively used for spatial visualization, overlay analysis, and suitability mapping. Emerging studies demonstrate the incorporation of AI and machine learning (ML) methods such as Improved whale optimization algorithm (IWOA), particle swarm optimization (PSO), and random forest to enhance site prediction accuracy, adaptability, and planning under uncertainty. Despite methodological progress, gaps remain in the consistent integration of social sustainability dimensions, dynamic real-time data, and empirical model validation. This review contributes by synthesizing current trends, identifying research limitations, and proposing a hybrid GIS–MCDM–AI framework to support future decision-making processes. The insights generated offer practical implications for infrastructure planners and policymakers aiming to scale up context-sensitive, intelligent, and sustainable EVCS deployment worldwide.