酒駕再犯之再犯及量刑因子分析與量刑最佳化

Abstract

酒後駕車一直是新聞媒體與社會密切關心的議題，政府也持續在推動修法與推行新政策，希望能確實達到嚇阻效果，降低國內酒駕發生率。然而在持續加重愈趨嚴格的法律規定約束下，也因為酒駕再犯情形還是不斷發生，而使社會大眾對法律在酒駕者上的實際懲處成效產生質疑。 本研究運用2010至2021年這11年間的裁判書，透過案件內容實證分析的方式，首先從中抓取重要資訊成為變數，接著運用Cox比例風險模型以及多元線性迴歸模型對數次酒駕再犯者的犯罪特性以及法官的判刑標準這兩個面向作深入探究，最後參考以上兩個面向的研究結果設計一個數學規劃模型，以能最大化酒駕再犯者的再犯時間間隔為目標求得刑罰判決組合的最佳解，期望提供法官判決參考，解決刑罰量刑過輕問題並有效嚇阻酒駕再犯。 從上述兩個面向的研究結果發現，男性的酒駕再犯風險高於女性，年齡層方面則是介於17至49歲的青壯年人口有相對高的酒駕再犯風險，此外，有酒駕肇事紀錄者或是教育程度較低下者也都是酒駕再犯的高風險族群。刑罰方面，相對於處以其他種類的刑罰，若對一直再犯的酒駕犯罪者處以刑期併科罰金，對降低他們酒駕再犯風險的幫助不大。而結果亦顯示酒駕酒精濃度越高的酒駕再犯者，他們的再犯風險也會隨之增高，這代表不論刑罰判的或輕或重，對於高酒精濃度酒駕者已無法造成任何影響，應對此族群採取其他種對策。

Drunk driving has been a great concern to the society for years, while the government has continually dedicated to the amendments of drunk driving law and the implementation of new policies, hoping to achieve a deterrent effect and have a decline on the incidence of drunk driving in Taiwan. However, under the restriction of the law that become harsher, the recurrence of drunk driving makes the public have doubt about the law’s effect on drunk driving. The data used in this study is court judgement documents from 2010 to 2021, By means of empirical analysis of the content of the case, we first transformed some important information into factors. Then, by using Cox proportional hazard model and multivariate linear regression model, this study dived into the criminal characteristics of drunk driving recidivists and the sentencing criteria of judges. Lastly, a mathematical programming model was designed with reference to the research results from the above two models. With the goal of maximizing the time interval until the next time re-offending of drunk driving, the best solution for the combination of sentence would be obtained. It is hoped that it will provide the sentence suggestions to the judge, so as to solve the problem of improper sentencing and effectively deter the repeat occurrence of drunk driving. From the results of the study, Men have a higher risk of re-offending while young adults aged between 17 and 49 have a relatively high risk of re-offending. In addition, those have caused accident while drunk driving or those with lower educational level are also at high risk of reoffense and committing DUI crime again. In terms of penalties, the imposition of sentences and fines at the same time on those who have been recidivists of drunk driving, as opposed to other types of penalties, is of little help in reducing their risk of recidivism of drunk driving. From the results can we also knew that the higher the alcohol concentration of drunken driving re-offenders, their risk of re-offending would also increase, which means that regardless of the severity of the sentence, it won’t cause any effect on the drunken drivers when they have a high alcohol concentration. In this case, other measures should be taken to against this group of people.