近年來，借助於電子科技的發展，高速公路建置多元的交通偵測設備，使得即時交通資訊的取得更加容易。這些交通資訊不僅能提供用路者旅行時間預測、交通路況警示、以及路徑導引等相關訊息，也讓高速公路管理單位能夠採取主動式的管理策略，以提升高速公路的行車安全與服務效率。
自從2014年電子收費系統全面開通以來，電子收費系統不僅提升用路人的方便性與行車效率，也提供管理單位穩定的即時交通資訊。線圈偵測器(VD)與電子收費系統(ETC)取得的交通資訊具有不同的性質，但皆能成為建立事故預測模型的參數，用來識別具有風險的路況，作為預先示警與主動控制策略的參考依據。本研究擬針對最近四年的高速公路交通事故建立預測模型，以國道一號與國道三號為建模的對象。依據國道警察局事故紀錄，從高速公局的開放資料庫擷取對應的交通資訊進行分析。事故預測採用條件羅吉斯迴歸模型，該模型可以控制干擾變數的影響，並分析交通參數對於事故概率的影響。
研究結果顯示，ETC預測模型中的平均流量以及平均重車比，以及VD預測模型中的流量標準差與佔有率的變異係數為顯著的事故預測子。相同偵測器不同路段預測模型中，顯示國道一號與三號具有不同的交通流狀況，因此建議分別建立事故預測模型。固定時段的預測模型中，由於VD模型可以納入交通流的變異，因此大部分VD模型比ETC有較佳的預測結果。在比較兩個不同時段的預測能力時，接近事故發生時間的時段顯示有較佳的預測性能，但在實務應用上，還需考量交通控制策略的有效影響時間來做決策。

In recent years, the development of new technologies for traffic surveillance system on freeways has provided multiple access to traffic flow data. These data can be informative for road users by providing travel time estimation, route guidance, and alarm of abnormal traffic states, and be applied by the freeway agency for Active Traffic Management (ATM) strategies to enhance the safety and the service level of freeway systems.
The primary objective of this study is to establish suitable crash prediction models with real-time traffic flow data extracted from Vehicle Detectors (VD) and Electronic Toll Collection (ETC) system to improve the freeway safety by taking precautions. The selected network consists of two main freeway sections in Taiwan, which are National Freeway No.1 (N-1) and National Freeway No.3 (N-3). With the crash records from National Highway Policy Bureau (NHPB), this study matches the crash data with the corresponding traffic data from Taiwan Area National Freeway Bureau (TANFB). The conditional logistic regression analysis is estimated to evaluate the impact of traffic variables while controlling the effect of confounding factors.
The empirical results indicate that the average volume and the average heavy car ratio in ETC based models and the standard deviation of volume and the coefficient of variation of occupancy in VD based models are significant crash precursors. Besides, separate models should be established on N-1 and N-3 since they have different traffic flow conditions. In terms of time specific prediction models, the prediction performance of VD based models is slightly better than ETC based models, for VD based models can present more details of traffic characteristics. The comparison of prediction models for different time slices shows the models with the time slice near the crash time have better prediction performance, while the effective time of traffic control strategies should be considered in practice.

1. Abdel-Aty, M., Dilmore, J., & Dhindsa, A. (2006). Evaluation of variable speed limits for real-time freeway safety improvement. Accident Analysis and Prevention, 38(2), 335-345. doi:10.1016/j.aap.2005.10.010
2. Abdel-Aty, M., & Pande, A. (2005). Identifying crash propensity using specific traffic speed conditions. J Safety Res, 36(1), 97-108. doi:10.1016/j.jsr.2004.11.002
3. Abdel-Aty, M., Uddin, N., Pande, A., Abdalla, M. F., & Hsia, L. (2004). Predicting freeway crashes from loop detector data by matched case-control logistic regression. Statistical Methods and Safety Data Analysis and Evaluation, 1897(1897), 88-95. doi:10.3141/1897-12
4. Abdel-Aty, M. A., & Pemmanaboina, R. (2006). Calibrating a real-time traffic crash-prediction model using archived weather and ITS traffic data. IEEE Transactions on Intelligent Transportation Systems, 7(2), 167-174. doi:10.1109/Tits.2006.874710
5. Agresti, A. (2007). An Introduction to Categorical Data Analysis: Wiley.
6. Ahmed, M., & Abdel-Aty, M. (2013). A data fusion framework for real-time risk assessment on freeways. Transportation Research Part C-Emerging Technologies, 26, 203-213. doi:10.1016/j.trc.2012.09.002
7. Ahmed, M., & Abdel-Aty, M. (2012). The viability of using automatic vehicle identification data for real-time crash prediction. IEEE Transactions on Intelligent Transportation Systems, 13(2), 459-468. doi:10.1109/Tits.2011.2171052
8. Council, N. R. (2010). HCM 2010 : highway capacity manual. Washington, D.C: Fifth edition. Washington, D.C. : Transportation Research Board, c2010-.
9. Golob, T. F., Recker, W. W., & Alvarez, V. M. (2004). Freeway safety as a function of traffic flow. Accident Analysis and Prevention, 36(6), 933-946. doi:10.1016/j.aap.2003.09.006
10. Hassan, H. M., & Abdel-Aty, M. A. (2013). Predicting reduced visibility related crashes on freeways using real-time traffic flow data. J Safety Res, 45, 29-36. doi:10.1016/j.jsr.2012.12.004
11. Hourdos, J., Garg, V., & Michalopoulos, P. (2008). Accident Prevention Based on Automatic Detection of Accident Prone Traffic Conditions: Phase I.
12. Hourdos, J. N., Garg, V., Michalopoulos, P. G., & Davis, G. A. (2006). Real-time detection of crash-prone conditions at freeway high-crash locations. Artificial Intelligence and Advanced Computing Applications, 1968(1968), 83-91. doi:10.3141/1968-10
13. Huang, Z., Gao, Z., Yu, R., Wang, X., & Yang, K. (2017, 24-26 May 2017). Utilizing latent class logit model to predict crash risk. Paper presented at the 2017 IEEE/ACIS 16th International Conference on Computer and Information Science (ICIS).
14. Kwak, H. C., & Kho, S. (2016). Predicting crash risk and identifying crash precursors on Korean expressways using loop detector data. Accident Analysis and Prevention, 88, 9-19. doi:10.1016/j.aap.2015.12.004
15. Lee, C., Abdel-Aty, M., & Hsia, L. (2006). Potential real-time indicators of sideswipe crashes on freeways. Safety Data, Analysis, and Evaluation, 1953(1953), 41-49. doi:10.3141/1953-05
16. Lee, C., Saccomanno, F., & Hellinga, B. (2002). Analysis of crash precursors on instrumented freeways. Statistical Methodology: Applications to Design, Data Analysis, and Evaluation, 1784(1784), 1-8. doi:Doi 10.3141/1784-01
17. Mirshahi, M., Obenberger, J., Fuhs, C. A., Howard, C. E., Krammes, R. A., Kuhn, B. T., . . . Yung, J. L. (2007). Active Traffic Management: The Next Step in Congestion Management.
18. Oh, C., Oh, J., & Ritchie, S. (2005). Real-time hazardous traffic condition warning system: framework and evaluation. IEEE Transactions on Intelligent Transportation Systems, 6(3), 265-272. doi:10.1109/tits.2005.853693
19. Pande, A., Abdel-Aty, M., & Hsia, L. (2005). Spatiotemporal variation of risk preceding crashes on freeways. Statistical Methods; Highway Safety Data, Analysis, and Evaluation; Occupant Protection; Systematic Reviews and Meta-Analysis, 1908(1908), 26-36. doi:10.3141/1908-04
20. Roshandel, S., Zheng, Z., & Washington, S. (2015). Impact of real-time traffic characteristics on freeway crash occurrence: systematic review and meta-analysis. Accident Analysis and Prevention, 79, 198-211. doi:10.1016/j.aap.2015.03.013
21. Shi, Q., & Abdel-Aty, M. (2015). Big Data applications in real-time traffic operation and safety monitoring and improvement on urban expressways. Transportation Research Part C: Emerging Technologies, 58, 380-394. doi:10.1016/j.trc.2015.02.022
22. Shi, Q., Abdel-Aty, M., & Lee, J. (2016). A Bayesian ridge regression analysis of congestion's impact on urban expressway safety. Accident Analysis and Prevention, 88, 124-137. doi:10.1016/j.aap.2015.12.001
23. Sun, J., & Sun, J. (2015). A dynamic Bayesian network model for real-time crash prediction using traffic speed conditions data. Transportation Research Part C-Emerging Technologies, 54, 176-186. doi:10.1016/j.trc.2015.03.006
24. Tanishita, M., & van Wee, B. (2017). Impact of vehicle speeds and changes in mean speeds on per vehicle-kilometer traffic accident rates in Japan. IATSS Research, 41(3), 107-112. doi:10.1016/j.iatssr.2016.09.003
25. Theofilatos, A., & Yannis, G. (2014). A review of the effect of traffic and weather characteristics on road safety. Accident Analysis and Prevention, 72, 244-256. doi:10.1016/j.aap.2014.06.017
26. Wang, C., Quddus, M. A., & Ison, S. G. (2013). The effect of traffic and road characteristics on road safety: A review and future research direction. Safety Science, 57, 264-275. doi:10.1016/j.ssci.2013.02.012
27. Wang, L., Abdel-Aty, M., Shi, Q., & Park, J. (2015). Real-time crash prediction for expressway weaving segments. Transportation Research Part C-Emerging Technologies, 61, 1-10. doi:10.1016/j.trc.2015.10.008
28. Xu, C., Liu, P., Wang, W., & Li, Z. (2012). Evaluation of the impacts of traffic states on crash risks on freeways. Accident Analysis and Prevention, 47, 162-171. doi:10.1016/j.aap.2012.01.020
29. Xu, C., Tarko, A. P., Wang, W., & Liu, P. (2013). Predicting crash likelihood and severity on freeways with real-time loop detector data. Accident Analysis and Prevention, 57, 30-39. doi:10.1016/j.aap.2013.03.035
30. Xu, C., Wang, W., Liu, P., & Li, Z. (2015). Calibration of crash risk models on freeways with limited real-time traffic data using Bayesian meta-analysis and Bayesian inference approach. Accident Analysis and Prevention, 85, 207-218. doi:10.1016/j.aap.2015.09.016
31. Yu, R., & Abdel-Aty, M. (2013a). Multi-level bayesian analyses for single- and multi-vehicle freeway crashes. Accident Analysis and Prevention, 58, 97-105. doi:10.1016/j.aap.2013.04.025
32. Yu, R., & Abdel-Aty, M. (2013b). Utilizing support vector machine in real-time crash risk evaluation. Accident Analysis and Prevention, 51, 252-259. doi:10.1016/j.aap.2012.11.027
33. Yu, R., Wang, X., & Abdel-Aty, M. (2017). A hybrid latent class analysis modeling approach to analyze urban expressway crash risk. Accident Analysis and Prevention, 101, 37-43. doi:10.1016/j.aap.2017.02.002
34. Zheng, Z., Ahn, S., & Monsere, C. M. (2010). Impact of traffic oscillations on freeway crash occurrences. Accident Analysis and Prevention, 42(2), 626-636. doi:10.1016/j.aap.2009.10.009