This paper examines the impact of data flows on different message delivery methods in DTN. We design an enhanced recovery method specifically for vehicular networks operating within grid road scenario. Our approach optimizes storage utilization by enabling intermediate nodes to generate anti-packets. This method proactively manages data flow, ensuring efficient handling of messages before they reach their final destinations. In this work, we compare the performance of the enhanced recovery method considering two widely recognized DTN protocols: Spray and Wait (SpW) and Epidemic. The evaluation focuses on data frame, acknowledgment, message delivery rate, delay, and network overhead metrics under various conditions. The simulation results have demonstrated that the enhanced recovery method effectively reduces acknowledgment and frame errors while maintaining a high delivery ratio. Additionally, ER-SpW significantly improves delay performance compared to conventional SpW, and ER-Epidemic provides stable reliability.