根據世界衛生組織（WHO）的統計，全世界每年有超過125萬人死於道路交通事故，且有約2-5千萬人遭受非致命傷害。有鑑於此，如何減少道路交通事一直以來都是各國的關注的安全議題之一。因為交通事故的發生不單僅是造成人員傷亡，還會額外給社會帶來其他更多問題及社會成本。由於交通安全事故的頻繁與嚴重性，我政府也致力進行道路交通事故的減低；然而，台灣貨運業目前仍沒有適當的安全評估機制。在這項研究中，我們結合了美國FMCSA的CSA程序（FMCSA，2017），日本的租賃巴士安全評估認證以及台灣的安全評量機制，為台灣三種不同的貨運業(汽車貨運業，汽車路線貨運業和汽車貨櫃貨運業)分別建立了新的合適的安全評量系統。此研究中，我們設計了一個兩階段的決策方法來建構的貨運業安全評量平台；第一階段我們利用模糊層次分析法（Fuzzy AHP）建立各公司的初始安全分數及等級，第二階段使用多種機器學習方法對前述資料進行分析學習來調整安全屬性的權重，最後使用調整完的權重來計算每個公司的最終安全分數和排名，而計算出的最終安全分數及排名可以供政府及民眾參考，政府可判斷貨運公司是否安全上有疑慮、是否有立即輔導之必要；而民眾可藉由此數據選擇較安全之貨運公司，以保障其貨物運送之安全。

According to the statistics from World Health Organization (WHO), there are over 1.25 million people die from road traffic accidents in the world each year, and about 20-50 million people suffering non-fatal injuries. In view of this, how to reduce road traffic accidents has become as one of the main improvement projects in various countries. In addition, the occurrence of traffic accidents not only brings casualties to passers-by, but also causes other problems and costs to society. Due to the frequency and severity of traffic safety accidents, our government is committed to reducing road traffic accidents. However, there is still no proper safety evaluation mechanism for the freight industry in Taiwan. In this research, we combine CSA program of FMCSA in the USA (FMCSA, 2017), Renting Bus safety evaluation in Japan, and some of safety evaluation in Taiwan to establish a new and suitable safety evaluation system for this three different freight industry in Taiwan. We designed a two-stage decision making approach to develop this mechanism. First, we apply fuzzy Analytic Hierarchy Process (AHP) to establish the initial safety rank of each company. After that, we use machine learning methods to analyze the data and adjust the weights of the safety attributes in the second. At last, we use the weights generated by two stage above to calculate each company’s safety performance and ranking. And these rankings can be used by the government or the public to judge whether the safety attributes of each company meet the regulations.

Breiman, L. (2001a). Random forests. Machine Learning, 45(1), 5-32.
Breiman, L. (2001b). Statistical modeling: The two cultures (with comments and a rejoinder by the author). Statistical Science, 16(3), 199-231.
Buckley, J. J. Fuzzy hierarchical analysis. Fuzzy Sets and Systems, 17(3), 1985, pp.233-247.
Chen, C.H. (2018). Taiwan Passenger Transportation Industry's Safety Evaluation Mechanism: Using Machine Learning Based Approach.
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3),
273-297.
Chen, T., Guestrin,. C. (2016). XGBoost: A Scalable Tree Boosting System. KDD '16: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785-794.
Chen, Z., Jiang, F., Cheng, Y., Gu, X., Liu, W., Peng, J. (2018). XGBoost Classifier for DDoS Attack Detection and Analysis in SDN-based Cloud. IEEE International Conference on Big Data and Smart Computing: 251-256.
Chen, C.H. (2018). Taiwan Passenger Transportation Industry's Safety Evaluation Mechanism: Using Machine Learning Based Approach.
Dong, L.-j., Li, X.-b., & Kang, P. (2013). Prediction of rockburst classification using Random Forest. Transactions of Nonferrous Metals Society of China, 23(2), 472-477.
Edwards, J.R.D., Davey,J., & Armstrong,K.A. (2014).Profiling contextual factors which influence safety in heavy vehicle industries. Accid.Anal.Prev.73,340-350.
Ho, T. K. (1995). Random decision forests. Paper presented at the Document analysis and recognition, 1995., proceedings of the third international conference on.
Hsu, W. K., Huang, S.H., & Tseng, W.J. (2016). Evaluating the risk of operational
safety for dangerous goods in airfreights - A revised risk matrix based on fuzzy AHP. Transportation Research, Part D 48, 235-247.
Huang, S., Cai, N., Pacheco, P. P., Narandes, S., Wang, Y., & Xu, W. (2018). Applications of Support Vector Machine (SVM) Learning in Cancer Genomics. Cancer Genomics-Proteomics, 15(1), 41-51.
Mollicone, D., Kan, K., Mott, C., Bartels, R., Bruneau, S., & Van Wollena, M. (2019). Predicting performance and safety based on driver fatigue. Accident Analysis and Prevention 126,142-145.
Morrow, P. C., & Crum, M. R. (2004). Antecedents of fatigue, close calls, and crashes among commercial motor-vehicle drivers. Journal of Safety Research, 35(1), 59-69.
Nguyen, A.T., Nguyen, L.D., Le-Hoai, L., & Dang, C.N. (2015). Quantifying the
complexity of transportation projects using the fuzzy analytic hierarchy process. International Journal of Project Management 33, 1364-1376.
Ortalda, A., Moujahid, A., Hina, M.D., Soukane, A. & Ramdane-Cherif, A. Safe Driving Mechanism: Detection, Recognition and Avoidance of Road Obstacles. In Proceedings of the 10th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management - Volume 2, 96-107.
Rodriguez-Galiano, V., Sanchez-Castillo, M., Chica-Olmo, M., & Chica-Rivas, M. (2015). Machine learning predictive models for mineral prospectivity: An evaluation of neural networks, random forest, regression trees and support vector machines. Ore Geology Reviews, 71, 804-818.
Saaty, T. L. (2008). Decision making with the analytic hierarchy process. International Journal of Services Sciences, 1(1), 83-98.
Sangwook, L. Application of AHP and Fuzzy AHP to Decision-Making Problems in
Construction. 52nd ASC Annual International Conference Proceedings.
Sun, D., Jia, Y., Wu, J., Chen, J., & Zhao, L. (2018). Application of fuzzy theory in the
optimization algorithm of the indicator weight for railway passenger transport safety evaluation. Intelligent Decision Technologies 12, 483-490.
Sparrow, AR., Mollicone, D., Kan, K., & Bartels, R. (2016). Naturalistic field study of the restart break in US commercial motorvehicle drivers: Truck driving, sleep, and fatigue. Accident Analysis and Prevention 93,55-64.
Peng, Z., Gao, S., Li, Z., Xiao, B., Qian, Y. (2018). Vehicle Safety Improvement through
Deep Learning and Mobile Sensing. IEEE Network • July/August 2018, 28-33.
Chang, Y.C., Chang, K.H., Wu, G.J. (2018). Application of eXtreme gradient boosting trees in the construction of credit risk assessment models for financial institutions; Applied Soft Computing Journal 73: 914–920.
Zhang, G., Yau, K., Zhang, X., & Lia, Y. (2016). Traffic accidents involving fatigue driving and their extentof casualties. Accident Analysis and Prevention 87, 34-42.
Zadeh, L. A. (1996). Fuzzy sets Fuzzy Sets, Fuzzy Logic, And Fuzzy Systems: Selected Papers by Lotfi A Zadeh : World Scientific, 394-432.
Žylius, G. (2017). Investigation of Route-Independent Aggressive and Safe Driving Features Obtained from Accelerometer Signals. IEEE Intelligent transportation systems magazine (summer),103-113. 24.
108年貨運三業EIS管理計畫及考核 (2019，11月19日)。
https://khcmv.thb.gov.tw
日本公益法人日本巴士協會租賃巴士安全評估認證網站。http://www.bus.or.jp/safety/about.html
國道交通事故統計及特性分析 (2019，11月14日)。
https://www.freeway.gov.tw/Publish.aspx?cnid=516&p=2849