道路交通事故衍生車輛延滯﹑能耗及污染排放之推估模式

Abstract

道路交通事故不僅造成人員死傷、財物損失等內部性成本外，還導致了如：額外車輛延滯、能源消耗及汙染排放等外部性成本。過往有多數研究注重於推估交通事故的內部性成本，推估交通事故外部性成本影響的研究則偏少。因此，本研究以道路交通事故發生的非重現性延滯為主軸，建構一整合模式推估因道路交通事故對上游車流產生衝擊而造成的額外車輛延滯與其產生的額外能耗與污染氣體排放。 此道路交通事故衝擊分析推估模式使用系統化的概念，以不同道路型態、事故衝擊範圍作為分類，結合事故延時模式、道路容量縮減設定模式、事故延滯模式、能源消耗模式與污染排放模式等來進行事故影響推估。本研究採用常見的等候、衝擊波等兩種車流理論方法，比較兩方法下推估結果的異同。另外，本研究嘗試了新穎路口占用推算法，來準確估計市區道路下路口事故占用不同區域時，對道路容量縮減之影響。 在模式的實際應用方面，本研究收集我國2011年高速公路事故資料、2012年臺北市市區道路事故資料、高速公路上道路偵測器的巨觀車流資料及臺北市的流量調查資料。模式應用推估結果顯示，「事故上游車流量」、「事故延時長度」與「道路占用情況」是影響推估結果的重要變數。另外，雖然兩種車流理論方法於延滯時間推估應用下，延滯推估的結果相差不大，但等候理論模型推估出的能源消耗量與污染排放量則皆低於衝擊波模型。推其原因為，在等候理論模型下，假設車輛延滯皆是以停等延滯的型式，只採用了怠速能耗係數與怠速碳排係數來推估事故造成的衍生影響。然而，衝擊波模型增加考量了行經事故路段前後的速率變化，而採用了隨速率改變的能耗係數與碳排係數，因此可推估出較準確的能源消耗量與污染排放量等事故衍生影響。

Road traffic accidents cause not only internal costs of fatalities, injuries and property damage, but also the external costs, such as delays, energy consumption and emissions. Numerous studies have been conducted to estimate the internal cost of accidents. However, few studies further attempt to estimate the external cost. Based on this, this study aims to propose traffic flow models to estimate accident-induced delays, energy consumption and emissions. The proposed model integrates traffic flow, energy consumption and emission models with the consideration of accident duration and capacity reduction of various types of accidents. Two traffic flow models are adopted and compared, including the queuing theory and the shockwave model. Additionally, to accurately estimate the capacity reduction at intersections, a novel method to estimate the capacity reduction according to the areas blocked is proposed. For field investigation and application, accident data along with corresponding traffic flow and geometric data in freeway systems and in local roadway systems in Taipei are respectively collected and field applied. The results show that traffic flow, accident duration, and capacity reduction are key variables for estimating abovementioned external negative effects. It is also found that although the total delays estimated by the two traffic flow models are exactly the same, but the total energy consumption and emissions estimated by the queuing model are substantially lower than those of the shockwave model. It is because that the queuing model treats all delays as stopped delay and only idle fuel consumption coefficient is used to estimate energy consumption and emissions; however, the shockwave model further traces the dynamic changes in travel speed in response to accidents (i.e. capacity reduction) and provides a better estimation for energy consumption and emissions.