Critical factors influencing consumer purchases of new energy vehicles: Evidence from a hybrid simulation model and machine learning

Abstract

New energy vehicles are distinguished by their environmentally friendly characteristics; however, their sales remain low compared with those of fossil fuel vehicles. This study identifies the key factors influencing consumer purchases of new energy vehicles with the aim of increasing sales and promoting the development of the new energy vehicle industry. Machine learning (ML) techniques were employed to analyze more than 26 000 consumer reviews of new energy vehicles and identify the factors influencing their purchasing decisions. Subsequently, a hybrid simulation model was constructed using the complex adaptive system framework. The model combines the strengths of ML, multi-agent modeling, and system dynamics modeling, thereby enhancing its capacity to identify key factors. The model comprises two categories of manufacturers: producers of new energy vehicles and producers of fossil fuel vehicles. Dynamic simulations were conducted using the AnyLogic simulation platform. The findings indicated a negative correlation between consumer income and demand for new energy vehicles. Conversely, expansions in the interior space of new energy vehicles and favorable consumer perceptions of their environmental benefits have contributed to increased demand for these vehicles. Therefore, rather than exclusively focusing on price advantages, it would be more beneficial to prioritize expanding new energy vehicles’ interior space and disseminating information about their environmental benefits to consumers. It seems reasonable to posit that these strategies would increase consumer willingness to purchase new energy vehicles in the long term.