A Novel Approach to Taxi-GPS-Trace-Aware Bus Network Planning

Taxi GPS traces are rich with information regarding the human mobility pattern in metropolitans. In this paper, we aimed at fully exploiting the Taxi GPS traces and addressing the bus network planning problem. Specifically, the proposed framework comprises a method for determining candidate bus stations by utilizing passenger pick-up and drop-off records, a bio-inspired method for yielding bus routes and further for generating the final bus network. To prove the effectiveness of our framework, we conduct simulative studies as well based on a real-world taxi GPS data-set and show that our proposed framework considerably outperforms traditional ones.