應用二元羅吉斯迴歸於高速公路重型車輛事故嚴重程度研究;A Study of Heavy Vehicle Accident Severity on Freeway Using Binary Logistic Model

NCUIR > Engineering College > Graduate Institute of Civil Engineering > Electronic Thesis & Dissertation > Item 987654321/87145

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/87145

Title:	應用二元羅吉斯迴歸於高速公路重型車輛事故嚴重程度研究;A Study of Heavy Vehicle Accident Severity on Freeway Using Binary Logistic Model
Authors:	林君翰;Lin, Jun-Han
Contributors:	土木工程學系
Keywords:	高速公路;肇事率;重型車輛事故;二元羅吉斯迴歸;Freeway;Accident rate;Heavy vehicle accident;Binary Logistic Regression
Date:	2021-08-25
Issue Date:	2021-12-07 15:08:09 (UTC+8)
Publisher:	國立中央大學
Abstract:	近年來台灣高速公路整體事故日漸增加，其中重型車輛事故增加約30%，重型車輛不僅有較高的肇事率，發生事故的嚴重程度也遠遠大於一般車輛，?能了解重型車輛事故嚴重程度之相關影響因素，對於高速公路事故預防或減少有相當程度的幫助。除此之外，有關於事故發生所造成死傷與社會成本損失，是重要的安全課題。本研究利用國道一號105-108年重型車輛事故原始資料，並使用二元羅吉斯迴歸針對事故嚴重程度進行分析，其影響因素經過蒐集整理分為車流、公路幾何、天候、事故背景以及其他特性，經由模式檢定找出顯著變數，探討重車事故在不同影響因素之嚴重程度。研究結果顯示，重車A1事故機率有正向影響之顯著變數為：速率、重車比例、風速、光線；反向影響之顯著變數為：流量、縱坡度、曲率半徑、車道數、飲酒情形、重型車輛種類、路面狀態。在不同公路型態情境下，調整可控制影響變數，當重型車輛事故發生，能有效降低 A1事故機率至0.00000022。;The overall number of freeway accidents in Taiwan has been increasing in recent years, among which heavy vehicle accidents have increased about 30%. Heavy vehicles not only have a higher rate of accidents, but the severity of accidents is far greater than that of ordinary vehicles. If you can understand the severity of heavy vehicle accidents related factors are helpful to prevention or reduction of freeway accidents. In addition, the death and injury caused by accidents and the loss of social costs are important safety issues. This study uses the original data of heavy vehicle accidents from 2016 to 2019 National No.1 Freeway, by using binary logistic regression to analyze the severity of accidents. The related factors are collected and sorted into traffic flow, freeway geometry, weather, accident background, and other characteristics. Through model verification, find significant variables, and explore the severity of heavy vehicle accidents in different related factors. The results show that the significant variables that have a positive impact on the probability of the heavy vehicle A1 accident are: speed, proportion of heavy vehicles, wind speed, and light; significant variables that have a negative impact are: traffic, longitudinal slope, radius of curvature, number of lanes, drinking situation. For the types of heavy vehicles and road conditions, adjusting the controllable variables under different road conditions can effectively reduce the probability of A1 accidents to 0.00000022 when heavy vehicle accident occurs.
Appears in Collections:	[Graduate Institute of Civil Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	108	View/Open

社群 sharing

Loading...