Key points are not available for this paper at this time.
The growing adoption of electric vehicles (EVs) depends on well-planned charging infrastructure. However, many EV charging station design models treat site allocation and charger configuration as separate processes, leading to spatial-capacity mismatches, infrastructure underutilization, and inefficient investment. To address this limitation, this study introduces an artificial intelligence (AI)-driven decision model that simultaneously optimizes station locations and on-site charger configurations. The model integrates two evolutionary AI techniques—genetic algorithm (GA) and differential evolution (DE)—into a hybrid metaheuristic framework. GA determines where to place charging stations, while DE decides the appropriate mix of DC fast and Type-2 chargers at each site. A case study in Detroit demonstrates the model’s effectiveness, revealing deployment patterns aligned with urban functions: residential and leisure zones require sparse fast-charging coverage; shopping zones benefit from dense fast-charging stations; and working zones are best served by dense Type-2 deployments. The model is computationally efficient and adaptable to varying urban contexts, offering a scalable and data-driven tool in charging station design for planners, engineers, policymakers, and industry stakeholders.
Hu et al. (Thu,) studied this question.