<p>This study delves into the competitive-cooperative relationship between shared e-bikes and buses, aiming to evaluate the influence of management policy changes on the travel choices of shared e-bike users and provide a basis for promoting their coordinated development. Shared e-bike travel orders are categorized based on the definition of the competitive-cooperative relationship. The MGWR model is utilized to analyze the spatial correlation between the built environment and the distribution of shared e-bike orders. A questionnaire survey is conducted to gather data on personal socioeconomic attributes, travel records, and psychological preferences, which are then integrated with the built environment data. The PSO-PNN model is employed to explore the impact of various factors on the competitive-cooperative relationship during different time periods and predict travel types. The research findings indicate differences in the factors influencing shared e-bike travel on weekdays and weekends. The PSO-PNN model outperforms the traditional binary logit model in prediction accuracy, achieving 91.80% on weekdays and 88.89% on weekends. In terms of policy evaluation, optimizing real-time bus arrival information, route design, and planning on weekdays can significantly increase the proportion of cooperative shared e-bike trips. On weekends, enhancing bus capacity and optimizing operations can raise the proportion of users choosing cooperative shared e-bike travel by up to 20%. The proposed approach enables policymakers to anticipate potential policy impacts before implementation. This study provides a quantitative foundation for developing data-driven transport policies and offers actionable insights to promote the coordinated development of shared e-bikes and buses.</p>