Factors Influencing Electric Vehicle Adoption in India: A Machine Learning Approach Using Artificial Neural Networks

Abstract

The adoption of electric vehicles (EVs) in India is influenced by a complex interplay of economic, infrastructural, policy-driven, and behavioural factors. This study employs Artificial Neural Networks (ANNs) to analyse survey data from 400 respondents to identify key determinants of EV adoption. The results indicate that cost-effectiveness and charging infrastructure availability are the most significant predictors of adoption, with cost concerns emerging as the most substantial barrier. Government incentives play a moderate role, while environmental consciousness and performance expectations have a comparatively lower influence. The ANN model achieved an accuracy of 88.2%, demonstrating high predictive power in assessing adoption likelihood. Sensitivity analysis further confirms that variations in cost perceptions and infrastructure availability have the greatest impact on adoption rates, emphasizing the need for targeted interventions. Policy recommendations suggest expanding financial incentives, improving charging networks, and structuring sustainable government subsidies. From an industry perspective, manufacturers must focus on cost reduction strategies, battery efficiency improvements, and expanding accessibility to charging infrastructure. This research provides valuable insights for policymakers, automakers, and stakeholders to accelerate the transition to sustainable mobility in India.