路邊停車場開單管理員交通受傷事件之研究 —以新北市為例—

Abstract

目前國內執行路邊停車收費作業，普遍採由開單管理員手持PDA數位化開單機器開單，且以步行或騎乘自行車、機車方式至收費路段。由於開單管理員長時間曝露於道路車流中，易生肇事。本研究透過對開單管理員交通受傷事件的分析與探討，瞭解其肇事原因與開單員受傷程度之相關性，期能有助於日後各級地方政府改善之參考。 本研究以新北市政府交通局94年6月至99年12月期間，登錄列管之113件開單管理員公傷事件進行探討。以皮爾森積差相關分析法（Pearson product-moment correlation）擇取具相關性顯著的肇事變數；再以卜瓦松迴歸方法，建立新北市路邊停車場開單管理員交通肇事最適模式。其中變數：「車道數」、「年齡」、「騎機車開單」、「肇事發生時段」、「肇事地點」等，與肇事發生次數呈現正相關。而「天候狀況」變數，則與肇事發生次數呈現負相關。研究結果發現開單管理員，年齡超過50歲、騎機車開單或於兩車道以下路段開單，肇事發生次數會顯著增加。 在影響開單管理員肇事受傷嚴重度分析部分，則以開單管理員所請公傷假天數區分2種受傷嚴重程度類別。經以皮爾森積差相關係數分析，與肇事受傷嚴重程度有顯著相關的變數，包含：「光線狀況」、「年齡」、「肇事型態」、「是否緊臨停車」、「肇事地點」等五個變數。再以二項式羅吉斯迴歸模式分析，得到結果為「年齡」、「光線狀況」、「是否緊臨停車」、「肇事型態」四個變數，與肇事受傷嚴重程度有顯著相關。研究結果發現，開單管理員面臨，年齡超過50歲、路邊有障礙物、於白天正常光線下執勤或執勤行經路口，一旦肇事發生，須請超過5天公傷假的機會顯著。

Currently in Taiwan, on-street parking fee issuing is commonly conducted by patrolmen, issuing fee notices with PDAs. These patrolmen walk or ride a bicycle or motorcycle to the sections where parking is charged. Accidents frequently happen to patrolmen due to their long exposure to traffic flows. This study analyzed and discussed the injury cases of patrolmen to understand the correlation between the causes and the degree of injury, hoping to offer local governments some effective ways to reduce injuries of patrolmen in the future. This study probed into 115 cases of patrolmen who had accidents on duty, recorded by the Transportation Department of New Taipei City Government. Pearson product-moment correlation was adopted to select representative injury cause variables. Poisson Regression Model was utilized to set up optimal model for traffic accidents of patrolmen in New Taipei City. The following variables are positively correlated to the occurrence of accidents: number of lanes, age of patrolmen, patrolling by a motorcycle, time intervals of accidents, and locations of accidents. On the other hand, weather conditions are negatively correlated to the occurrence of accidents. This research found that the possibility of occurrence of accidents would increase when patrolmen were aged over 50, or when patrolmen were riding motorcycles to issue fee notices, or when patrolmen were on the section which was less than two lanes. The degree of injury was classified into two types according to the number of days that patrolmen asked for leave. Analyzed with Pearson product-moment correlation, factors with significant correlation with the degree of injury are lightness, age of patrolmen, types of accidents, whether the patrolman was standing next to a parking space, and sites of accidents. Again, analyzed with Multinomial Logistic Regression, it showed the following four variables were significantly relevant to the degree of injury: age of patrolmen, lightness, whether the patrolman was standing next to a parking space, and types of accidents. The study also found once an accident happens, four conditions will raise the possibility of patrolmen’s asking for leave for over 5 days. The four conditions are patrolmen aged over 50, obstacle along the roadside, in common daylight and patrolmen passing intersections.