Agent-Based Modeling of Electric Vehicles Adoption: Assessing Sustainable Mobility from a Multi-Dimensional Perspective

In recent decades, significant attention has been paid to air pollution and climate change, focusing on the centrality of mobility that plays a fundamental role in the emission of greenhouse gases. As a result, there is growing attention toward the electrification of road transport to achieve environmental sustainability (Kumar & Alok, 2020). The deployment of the adoption of electric vehicles seems to be an effective strategic policy in this direction. Therefore, scholarly research on this topic has witnessed a substantial increase in the current decade; however, such literature is mainly focused on different subsets of dimensions characterizing the phenomenon, leaving unexplored areas that comprehensively integrate the economic, technical, and social-behavioural elements of analysis, and delve into how all the corresponding variables interact with each other within a unique framework. In this paper, we propose an agent-based model (ABM) to investigate the adoption of electric vehicles (EVs) in a community of individuals. ABM is particularly indicated for modelling complex systems when the agents are affected by heterogeneity and not linear interactions (Cozzoni et al., 2021). Thus, we simulate the diffusion of adoption of EVs considering simultaneously the effects of technical aspects (driving range autonomy, purchasing price, operating costs, and the diffusion of charging infrastructure), economic incentives and social influence (financial support and advertising), and behavioural elements (range anxiety and environmental awareness). The interactions between agents occur along different topological structures (we examine Small-World and Scale Free network topologies) in which they act according to Kowalska-Pyzalska's model (Kowalska-Pyzalska, 2017). The model is partially calibrated using a sample of Italian consumers (Giansoldati et al., 2018a). Several simulative scenarios were implemented, varying the value of the variables involved. The analysis of the configurations showed some preliminary results, i.e., the large-scale adoption of EVs can be achieved by reducing range anxiety and increasing environmental awareness without using external forces such as financial supports and promotion. Similarly, increasing the values of the variables travel autonomy and diffusion of charging infrastructure. It was also noted that the effectiveness of financial support depends on the configuration of behavioural variables. Finally, the model has been implemented in NetLogo 6.1.1. It represents a general model that, in the future, research advancements could be specified for empirical cases and policy advice interventions.