公車路網即時滯站與速度控制策略之最佳化與模擬研究

Optimization and Simulation of Real-time Holding and Speed Control Strategies in a Bus Network

班距不穩定性和串車現象常見於現實中公車運行狀況，往往會增加乘客額外等待時間。然而，滯站策略雖然被證明能有效改善服務穩定性，卻會額外造成乘客旅行時間成本提高。因此，本研究發展出即時滯站與速度控制策略，並建立一最佳化模式，其目標為在平面滾動法架構上最小化乘客總旅行和總等待時間。為滿足即時需求，本研究開發出啟發式求解演算法，能有效加速電腦計算效率。為驗證控制策略成效，本研究發展離散事件模擬系統，以模擬公車運行動態和乘客搭車過程。最後，本研究以小型案例與台北實際公車路網進行測試，結果顯示結合滯站策略與速度控制策略在短班距與高乘客需求之情境，能有效降低乘客等待時間與旅行時間，且適用於複雜之公車路網。

Instable headway and bus bunching are commonly observed in actual bus operation conditions, which often increase the extra waiting time of passengers. However, although the holding control has been proven to be effective in improving service stability, it will additionally increase the travel time of passengers. Therefore, this study developed an optimization-based strategy combined holding control and speed control, whose goal is to minimize the total travel time and total waiting time of passengers within a predicted rolling horizon. Due to the limit of real time, this study also built a heuristic method for the optimization problem, which can effectively accelerate the efficiency of calculation. In addition, in order to verify the performance of the control strategy, this study developed the discrete-event simulation system which can simulate the bus operation and movement process of passengers. some case studies and actual bus networks in Taipei were tested in this paper. The results show that the combination of holding control and speed control can effectively reduce passenger waiting time and travel time under the scenario of short headway and high demand, which is suitable for complex bus network.