分析第一當事者影響A1、A2類肇事事故因素之研究

Analysing the impact factors from the first party on fatal and injury accidents

黃顯欽(Xian-Qin Huang)

指導教授 ： 吳水威

國立交通大學/管理學院/運輸科技與管理學系/碩士(2011年)

https://doi.org/10.6842/NCTU.2011.00524

摘要

本研究針對A1類與A2類之交通事故中之第一當事者為研究對象，研究其影響肇事程度(A1、A2類)之重要因素。內容分為兩大部分，第一部分為蒐集內政部警政署民國92、93年道路交通事故之原始資料，以敘述性統計分析各變數所相對應之A1與A2類事故之肇事次數與A1/(A1+A2)比例，了解道路交通事故肇事程度之特性。第二部分為以卡方自動互動檢視法及卡方獨立性檢定找出影響肇事程度之重要因素，並以二元羅吉斯迴歸模型配適。
研究結果顯示事故發生時間方面，「凌晨五點」之勝算比為最高；道路類別方面，「國道」之勝算比為最高；事故類型及型態方面，「人與汽(機)車部分之在路上嬉戲」之勝算比為最高；主要肇事因素方面，「行人(或乘客)部分之未依規定行走行人穿越道、地下道、天橋而穿越道路」之勝算比為最高；保護裝備方面，「未戴安全帽或未安全帶(未使用幼童安全椅)」之勝算比為最高；飲酒情形方面，「經呼氣檢測未超過 0.25 mg/L或血液檢測未超過 0.05%」之勝算比為最高；車種方面，「曳引車」為最高。

關鍵字

交通事故、卡方自動互動檢視法、卡方獨立性檢定、二元羅吉斯迴歸

並列摘要

Accident severity concerned in this study is a dichotomous variable with two categories, fatal (A1) and injury (A2). The purpose of this study is to analysis the impact of factors from the first party on fatal and injury accidents. The content has two parts. First, this study use descriptive statistics to analysis each variable from accidents raw data of National Police Agency on the 2003 and 2004 by computing the number of A1 and A2 accidents, and the proportion of A1/(A1+A2). Based on the result, we can determine the property of accident severity of traffic accident. Second, we use Chi-Square Automatic Interactive Detector (CHAID) and Pearson Correlation Coefficient and test of independent to determine the impact factors of accident severity. Besides, we use Logistic regression to fit the model.
The results suggest that 5 o’clock in the morning, on the highway, pedestrians play on the road, pedestrians go through the road without following the rule, without safety hamlet and safety belt, and alcohol breath testing over 0.25 mg/L or alcohol blood testing over 0.05% are associated with highest odd ratio.

並列關鍵字

accidents, CHAID, test of independent, Logistic regression

參考文獻

22. 行政院衛生署，網址：

32. 許添本、李明聰，「地區性道路人車衝突交通安全風險評估模式之建立」，運輸計劃季刊，第三十一卷，第二期，pp.245-266，民國91年。

1. Abdel-aty, M. A., Chen, C. and Schott, J. R. (1998). “An Assessment of the Effect of Driver Age on Traffic Accident Involvement Using Log-linear Models”. Accident Analysis and Prevention, vol. 30, No. 6,pp. 851-861.

2. Abdelwahab, H. T.and Abdel-Aty, M. A. (2001). “Development of Artificial Neural Network Models to Predict Driver Injury Severity in Traffic Accidents at Signalizes Intersection“. Transportation Research Record. 1746, Paper No. 01-2234, pp. 6-13.

3. Al-Ghamdi, A. S. (2002). ” Using Logistic Regression to Estimate the Influence of Accident Factors on Accident Severity”. Accident Analysis and Prevention, vol. 34, No. 6, pp. 729-741.