交通號誌時制補償方法之研究

Abstract

道路網中的交叉路口是都市運輸系統中最重要的設施之一，其中更迭指派路權、引導道路車流、降低潛在肇事衝突等最迅速有效的方法是透過號誌系統來達成。號誌系統的控制績效與時制設計之良窳有密切的關連，有鑑於此，目前一般號誌控制器皆可建置多套時制計畫，能在預設的時間或有特定需求時進行時制轉換，以因應不同的交通狀況。由於目前很少實證顯示現有的時制轉換方法是否有效率，也無相關研究證實號誌時制的補償程序對於路口績效水準是否有顯著的影響，因此，當流量水準明顯改變而需進行時制轉換時，如何進行時制轉換與進行必要的時間補償作業，乃為本研究探討的重點。 除了參閱現有時制設計的各種要項與特性之外，本研究亦參考交通部運研所彙整的各種時制轉換方法，惟由各方法的特點與限制發現它們大都未由流量需求型態與變異的觀點著眼。因此，本研究首先探討在預定時制與均勻分配的需求型態下的基本補償模式，再由時制轉換點對於舊時制的分割型態，研擬兩種以需求漸變為導向的時制補償方法，並在滿足七個預設的控制條件之下，於新、舊時制之間加入一「轉換時段」，以便在此時段中針對前一時制被分割後部分或尚未執行的時相進行補償。 基於現場路口調查作業與操作號誌控制器功能的局部限制與安全考量，本研究以交通模擬整合系統 (TSIS) 為工具，對所發展的兩個時相補償法與彙總的其他方法論進行若干績效的分析與比較，同時亦針對新、舊時制之間不同的流量水準進行敏感度分析，顯示本研究發展的時制轉換法在不同的流量需求型態大都具有相對較佳的控制績效。

The intersections on a roadway network are one of the most critical facilities of an urban transportation system. Controlling through alternately assigning right-of-way, efficiently guiding traffic, and effectively reducing potential conflicts may be achieved by signal control systems. With multiple build-in timing plans, modern signal controllers have high flexibility to perform the function of timing switchovers at a given time or for a specific demand in order to adapt to various traffic conditions. Unfortunately, neither of the practical results showed that the existing approaches would operate effectively, nor of them proved that the compensation procedure for timing transitions might have significant impact to the traffic performance. Thus, as the timing adjustment is to be activated due to a dramatic change in demand, how to execute the policy of timing transition and phase compensation becomes an important issue that is worth while of exploring. Research on the operating characteristics, and limitation of timing transition methods between control periods has been reviewed by this study. It is seen that most of the methods were developed based on ideas without taking account of the flow patterns and their fluctuation trend. Thus, this study started with a basic formulation using the minimum cycle concept for the compensating phases in a presumed control horizon. Then, depending on the elapsed phases at the transient time, two demand based compensation approaches were proposed. With each of the seven control requirements, a “transition period” was defined in between the past and the incoming control timings so as to proceed the individual approach developed in this study. Due to the safety problems possibly incurred in setting and/or operating a signal controller at site, the FHWA’s TSIS was applied instead to analyze and compare the output performance of various approaches for the transition period. Sensitivity tests for different demand levels were also conducted to well catch the trend of different outputs. The analyses showed that the two newly developed approaches could lead to a relatively good performance as compared with the others under most of the demand levels.