With the steady advancement and application of smart transportation alongside accelerating lifestyles, the intelligent, high-quality, and efficient operation of public transit has become an increasingly critical societal concern. Recently, a research team led by Associate Professor Cao Zhichao from the School of Transportation and Civil Engineering proposed a predict-then-optimize modeling approach based on demand-responsive buses. By adopting a novel AI+OR collaborative methodology, the team transformed static demand allocation into real-time dynamic inputs with short-term forecasting.

This model successfully eliminated the professional barriers between real-time reservations and dynamic scheduling, effectively reducing passenger waiting times and significantly improving travel efficiency. The related research findings have been published in the internationally renowned academic journal Transportation Research Part E: Logistics and Transportation Review.

https://news.ntu.edu.cn/2025/0603/c8062a267118/page.htm