Houzhi Li

This study aims to evaluate the operational efficiency of electric vehicle (EV) charging stations and explore optimization strategies to enhance resource utilization and service performance. A systematic review approach was first applied to identify the main evaluation indicators and influencing factors from existing studies. Subsequently, a super-efficiency Data Envelopment Analysis (DEA) model was used to assess the efficiency of six EV charging stations in a certain City, China. The robustness analysis was carried out, and the output variables were replaced, and the evaluation results did not change. The results show substantial disparities in efficiency across stations: C1 exhibits the highest operational efficiency, while C3 performs the lowest. The inefficiencies primarily result from supply–demand mismatches and redundant capacity investment. Based on these findings, the study proposes both overall and localized optimization strategies to improve operational performance. The results provide valuable insights for urban energy infrastructure planning and contribute to the enhancement of high-quality, low-carbon transportation development in China.

User preferences are important for electric vehicle charging station (EVCS) recommendations, but they have not been deeply analyzed. Therefore, in this study, user charging preferences are identified and applied to EVCS recommendations using a hybrid model that integrates LightGBM and singular value decomposition (SVD). In the model, LightGBM is used to predict user ratings according to users’ comments regarding charging orders, and the feature importance reported by each user is output. Then, a co-occurrence matrix between users and charging stations (EVCSs) is constructed and decomposed using SVD. Based on the decomposed results, the final evaluated scores of each user for EVCSs can be calculated. Upon ranking the EVCSs according to the scores, the EVCS recommendation results are obtained, taking into account the users’ charging preferences. The sample data consist of 28,306 orders from 508 users at 241 charging stations in Linyi, Shandong, China. The experimental results show that the proposed hybrid model outperforms the benchmark models in terms of precision, recall, and F1 score, and its F1 score can be increased by 96% compared with that of the traditional item-based collaborative filtering method with charging counts for EVCS recommendations.