我國早期路口設計未充分考量左轉車流需求，部分臨近路口缺乏左轉專用車道配置，導致左轉車無法有效紓解，進而影響交通效率。隨著機動車數量逐年增加，路口交通衝突與事故問題亦日益嚴重，其中涉及左轉行為之穿越側撞、擦撞及追撞事故，往往造成較嚴重之人員傷害與財產損失。在現有道路空間限制下，道路主管機關多採用調整交通號誌時制試圖緩解左轉車流問題，最常使用的方式即是設置綠燈早開或遲閉時相，然而這些折衷方案反可能導致更多的交通衝突。由於國內對於這些時相設計缺乏明確的系統性研究和標準，而多依賴傳統或主觀決策，導致各個路口管制方式不一致，駕駛人難以遵循，恐增加混亂和事故發生機會，因此，本研究旨在探討在選擇實施綠燈早開或遲閉時，如何採取較為適當的時相配置方式，並提供基於我國道路環境的評估準則，以降低交通事故和衝突的風險，為相關單位提供設計參考。
本研究將以「安全」為出發點，以交通衝突角度進行分析，借助影像辨識技術分析左轉衝突型態，透過不同路口所具之車道配置、時相差異、轉向需求等環境因子的比較，以機器學習演算法和建立迴歸模型之方式推論交通衝突事件的發生與綠燈早開、遲閉時相之間可能存在的關聯性及影響趨勢，並嘗試分析不同左轉時相設計所連結的交通衝突態樣。
在彙整上述觀察結果後，本研究將針對不同左轉時相設計的安全性提出論證，並據以建立一套可供相關主管機關審酌、參考的決策指引，進而採行安全性較高的時制計畫方案。期望透過本研究的分析成果，能減緩左轉車流所帶來的困境，提升我國道路交通安全。

Early intersection designs in Taiwan underconsidered left-turn operational demand, resulting in insufficient capacity without exclusive left-turn lanes and reduced overall traffic efficiency. As motor vehicles increased, left-turn-related conflicts and crashes—such as crossing, sideswipe, and rear-end collisions—became increasingly severe, causing significant injuries and property losses.
Due to space constraints, local transportation agencies often rely on signal timing adjustments, such as leading or lagging green phases, rather than geometric improvements to enhance left-turn operations. However, these compromise measures may unintentionally generate additional conflicts or crashes under certain traffic conditions. Furthermore, standardized guidelines for these strategies remain fragmentary in Taiwan. Officials typically rely on subjective judgment or personal experience, leading to inconsistent signal control strategies even across similar intersections, which increases driver uncertainty and crash risks.
To address this issue, this study investigates appropriate implementation strategies for leading and lagging green phases and establishes evaluation criteria suitable for Taiwan’s roadway environment. Taking a traffic safety perspective, the study utilizes video-based trajectory recognition techniques to analyze traffic conflict patterns as surrogate safety measures. By comparing environmental factors—including lane configurations, signal designs, and turning demands—machine learning algorithms and regression models are employed to explore the relationships between traffic conflicts and leading or lagging green phases, while examining their associations with specific conflict types.
Based on comprehensive empirical results, this study provides evidence-based discussions on safety performance and proposes decision-making guidelines for signal timing selection. The findings are expected to support safer left-turn control strategies, mitigate left-turn traffic conflicts, and ultimately improve overall roadway safety in Taiwan.

Abdel-Aty, M. A., & Radwan, A. E. (2000). Modeling traffic accident occurrence and involvement. Accident; analysis and prevention, 32(5), 633–642.
Adamson, M., Schultz, G. G., Saito, M., & Stevens, M. D. (2020). Developing Decision Boundaries for Left-Turn Treatments. Transportation Research Record, 2674(5), 315-326. https://doi.org/10.1177/0361198120915470
Aguero-Valverde, J., & Jovanis, P. P. (2008). Analysis of Road Crash Frequency with Spatial Models. Transportation Research Record: Journal of the Transportation Research Board, 2061(1), 55-63. https://doi.org/10.3141/2061-07 (Original work published 2008)
Allen, Brian & Shin, B.Tom & Cooper, Peter. (1978). Analysis Of Traffic Conflicts And Collisions. Transportation Research Record. 67-74.
Baguley, C.J.(1984). The British Traffic Conflict Technique. Transport and Road Research Laboratory, NATO ASI Series, Vol F5, International Calibration Study of Traffic Conflict Techniques.
Bhim, Ravindranauth (2005). Observational Before and After Safety Study of Installing Signals at Rural Intersections: Using the Empirical Bayes (EB) and Conventional Methods. Toronto Metropolitan University. Thesis. https://doi.org/10.32920/ryerson.14649114.v1
Bo Lan, Bhagwant Persaud, Craig Lyon, Ravi Bhim(2009).Validation of a Full Bayes methodology for observational before–after road safety studies and application to evaluation of rural signal conversions, Accident Analysis & Prevention, Volume 41, Issue 3,,Pages 574-580.
Budzyński, A., Federowicz, M., Jabłoński, A., Hasan, W., Gorszanów, J., & Nikitishyn, T. (2025). A Machine Learning Approach for Predicting Road Accidents. Safety & Defense, 10(2), 60-70. https://doi.org/10.37105/sd.230
C. Gheorghe and N. Filip (2022). Road Traffic Analysis Using Unmanned Aerial Vehicle and Image Processing Algorithms. 2022 IEEE International Conference on Automation, Quality and Testing, Robotics (AQTR), Cluj-Napoca, Romania, 2022, pp. 1-5, doi: 10.1109/AQTR55203.2022.9802058.
Carriquiry, Alicia L.;Pawlovich, Michael(2004). From empirical Bayes to full Bayes : methods for analyzing traffic safety data. Iowa State University; Iowa. Dept. of Transportation.
Carson, Jodi L.,Powers, Meagan(2004). Before-After Crash Analysis: A Primer for Using the Empirical Bayes Method Tutorial. Montana State University (Bozeman, Mont.), Department of Civil Engineering. FHWA/MT-04-002-8117-21.
Chai Chen, Zeng Xianming, Wu Xiangbin, Wang Xuesong. (2020). Evaluation and Optimization of Responsibility-Sensitive Safety Models on Autonomous Car-Following Maneuvers. Transportation Research Record Journal of the Transportation Research Board. 2674. 10.1177/0361198120948507.
Davis, Gary. (2008). Bayesian Reconstruction of Traffic Accidents. Law Probability and Risk. 2. 10.1093/lpr/2.2.69.
Essa, M., & Sayed, T. (2019). Full Bayesian conflict-based models for real time safety evaluation of signalized intersections. Accident; analysis and prevention, 129, 367–381.
Gary A. Davis, Abhisek Mudgal(2013). Field Study of Driver Behavior at Permitted Left-Turn Indications. Intelligent Transportation Systems Institute, Center for Transportation Studies, University of Minnesota.
Gary A. Davis, Raphael Stern, Melissa Duhn, Jingru Gao(2023). Driver Comprehension of Flashing Yellow Arrows. Civil, Environmental, and Geo- Engineering University of Minnesota, Minnesota Department of Transportation Office of Research & Innovation. https://mdl.mndot.gov/
Gettman, D., Pu, L., Sayed, T. and Shelby, S. (2008) Surrogate Safety Assessment Model and Validation: Final Report, Federal Highway Administration, Technical report FHWA-HRT-08- 051.
Gjorgjievski, Marko & Khanna, V. & Petrisor, B. & Li, C.S. & Ristevski, B.. (2021). Roadside evaluation of distracted driving – Driver limitations in recognizing traffic light transitions (REDD – LIGHT). Journal of Transport & Health. 21. 101059. 10.1016/j.jth.2021.101059.
Glennon, J.C., Glaus, W.D., Sharp, M.C., Thorson, B.A.(1977). Critique of the Traffic-conflict Technique. Transportation Research Record 630, Washington, D.C.: Transportation Research Board.
Gongquan Zhang, Jieling Jin, Fangrong Chang, Helai Huang (2024). Real-time traffic conflict prediction at signalized intersections using vehicle trajectory data and deep learning, International Journal of Transportation Science and Technology.
Hauer, E., Harwood, D. W., Council, F. M., & Griffith, M. S. (2002). Estimating Safety by the Empirical Bayes Method: A Tutorial. Transportation Research Record, 1784(1), 126-131. https://doi.org/10.3141/1784-16
Hayward, J.C.. (1972). Near-miss determination through use of a scale of danger. Highway Research Record. 384. 24-34.
Heydari, M., & Amador-Jimenez, L. (2012). Comparing Full Bayes Likelihoods to Predict Road Accidents and Identify Potential Hazardous Sites. Journal of Civil Engineering and Science, 1 (3).
Hydén, C. (1987). The development of a method for traffic safety evaluation: The Swedish Traffic Conflicts Technique. Bulletin Lund Institute of Technology, Department(70).
Ismail Zohdy (2009). Modeling Permissive Left-Turn Gap Acceptance Behavior at Signalized Intersections. Virginia Tech. http://hdl.handle.net/10919/35691.
Jonathan Upchurch(1991). Comparison of Left-Turn Accident Rates for Different Types of Left-Turn Phasing. Transportation Research Record 1324,33-40.
Joseph E. Hummer, Robert E. Montgomery, Kumares C. Sinha(1989). An Evaluation of Leading Versus Lagging Left Turn Signal Phasing. Joint Highway Research Project FHWA/IN/JHRP-89/17.
Joseph E. Hummer, Robert E. Montgomery, Kumares C. Sinha(1991). Guidelines for Use of Leading and Lagging Left-Turn Signal Phasing. Transportation Research Record 1324,11-20.
Jovanis, P. P., & Chang, H.-L. (1986). Modeling The Relationship Of Accidents To Miles Traveled. Transportation Research Record, 42-51.
Kenneth R. Agent(1979). An Evaluation of Permissive Left-turn Phasing. Research Report 519 KYP-75-70; HPR-PL-1(14), Part Ill-B.
Laureshyn, A., Svensson, Å., & Hydén, C. (2010). Evaluation of traffic safety, based on micro-level behavioural data: Theoretical framework and first implementation. Accident Analysis & Prevention, 42(6), 1637-1646.
Laureshyn, Aliaksei & Varhelyi, Andras. (2018). The Swedish Traffic Conflict Technique - Observer's manual. https://portal.research.lu.se/en/publications.
Lawalata, G. M., & Agah, H. R. (2011). Traffic conflict analysis as a road safety diagnostic tool for urban road facilities. International Journal of Technology, 2(2), 112-121.
Lin Zhang, Panos D. Prevedouros, Honglong Li(2005). Warrants for Protected Left-Turn Phasing. ASCE Journal of Transportation Engineering.
Lin, F.B. (1992). Left-Turn Adjustment Factors For Saturation Flow Rates Of Shared Permissive Left-Turn Lanes. Transportation Research Record.
Mahmud, S. M. & Ferreira, L. & Hoque, Md & Hojati, Ahmad. (2018). Reviewing traffic conflict techniques for potential application to developing countries. Journal of Engineering Science and Technology. 13. 1869-1890.
Mahmud, S., Ferreira, L., Hoque, M.S., & Tavassoli, A. (2017). Application of proximal surrogate indicators for safety evaluation: A review of recent developments and research needs. Iatss Research, 41, 153-163.
Mao, Y., Wang, W. H., Guo, W. W., Ding, C. X., & Xia, S. C. (2012). Analysis of opposing left-turn conflicts based on traffic conflict technology. Journal of Beijing Institute of Technology, 21(4), 487-491.
Miaou, S.P., & Lum, H. (1993). Modeling vehicle accidents and highway geometric design relationships. Accident; analysis and prevention, 25 6, 689-709 .
Minderhoud, M.M., Bovy, P.H.L.(2001). Extended time-to-collision measures for road traffic safety assessment. Accident Analysis and Prevention 33, 89–97.
National Academies of Sciences, Engineering, and Medicine. (2015). Signal Timing Manual - Second Edition. Washington, DC: The National Academies Press.
National Academies of Sciences, Engineering, and Medicine.(2022). Highway Capacity Manual 7th Edition: A Guide for Multimodal Mobility Analysis. Washington, DC: The National Academies Press. https://doi.org/10.17226/26432.
Nikiforos Stamatiadis , Salee Tate , Adam Kirk(2006). Left-turn phasing decisions based on conflict analysis. Transportation Research Procedia 14 ( 2016 ),3390-3398.
Niture, N., Abdellatif, I. (2025). A systematic review of factors, data sources, and prediction techniques for earlier prediction of traffic collision using AI and machine Learning. Multimed Tools Appl 84, 19009–19037. https://doi.org/10.1007/s11042-024-19599-6
Obasi, I. C., & Benson, C. (2023). Evaluating the effectiveness of machine learning techniques in forecasting the severity of traffic accidents. Heliyon, 9(8), e18812. https://doi.org/10.1016/j.heliyon.2023.e18812
Perkins, S. R. (1968). Traffic conflict characteristics-accident potential at intersections. Highway Research Record, 225, 35-43.
Raessler A and Yang J (2017), Derivation of Decision Boundaries for Left Turn Treatments at Signalized Intersections. Transportation Research Record, No. 2620(1), pp.1-9.
Raghavan Srinivasan, Bo Lan, Daniel Carter, Sarah Smith, and Kari Signor(2020). Safety Evaluation of Flashing Yellow Arrow at Signalized Intersections. Office of Safety Research and Development and Federal Highway Administration FHWA-HRT-19-036.
Raju Thapa, Ph.D., P.E. (TX), Md Asaduzzaman, Kwabena Abedi(2022). Evaluating Permitted/Protected Versus Protected Left-turn Signals in Louisiana. Louisiana Department of Transportation and Development.
Richard van der Horst, Richard & de Goede, Maartje & Hair-Buijssen, Stefanie & Methorst, Rob. (2013). Traffic conflicts on bicycle paths: A systematic observation of behaviour from video. Accident; analysis and prevention. 62. 10.1016/j.aap.2013.04.005.
Seth A. Asante, Sramak A. Ardekani, and James C. Williams(1993). Selection Criteria for Left-Turn Phasing and Indication Sequence. Transportation Research Record 1421,11-20.
Shakil Ahmed, Md Akbar Hossain, Sayan Kumar Ray, Md Mafijul Islam Bhuiyan, Saifur Rahman Sabuj. (2023). A study on road accident prediction and contributing factors using explainable machine learning models: analysis and performance, Transportation Research Interdisciplinary Perspectives, Volume 19,100814,ISSN 2590-1982,https://doi.org/10.1016/j.trip.2023.100814.
Sheffer, C., & Janson, B. N. (1999). Accident and Capacity Comparisons of Leading and Lagging Left-Turn Signal Phasings. Transportation Research Record, 1678(1), 48-54. https://doi.org/10.3141/1678-07
Spiegelhalter, D.J., Best, N.G., Carlin, B.P. and Van Der Linde, A. (2002), Bayesian measures of model complexity and fit. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 64: 583-639. https://doi.org/10.1111/1467-9868.00353
Tarko, A. (2019) Measuring Road Safety with Surrogate Events. 1st Edition, Elsevier, Amsterdam, 1-252.
Van der Horst, R.(1990). A time-based analysis of road user behaviour in normal and critical encounters. Doctoral Thesis, Delft University of Technology.
Vogel K. (2003). A comparison of headway and time to collision as safety indicators. Accident; analysis and prevention, 35(3), 427–433. https://doi.org/10.1016/s0001-4575(02)00022-2
W.T. Baker(1971). An Evaluation of the Traffic Conflicts Technique. Federal Highway Administration.
WAKABAYASHI, Hiroshi & TAKAHASHI, Yoshihiko & NIIMI, Shigehiro & Renge, Kazumi. (2003). Traffic Conflict Analysis using Vehicle Tracking System/Digital VCR and Proposal of a New Conflict Indicator. INFRASTRUCTURE PLANNING REVIEW. 20. 949-956. 10.2208/journalip.20.949.
Wong SC, Sze NN, Li YC. Contributory factors to traffic crashes at signalized intersections in Hong Kong. Accid Anal Prev. 2007 Nov;39(6):1107-13. doi: 10.1016/j.aap.2007.02.009. Epub 2007 Mar 19. PMID: 17920832.
Xu, X., Wang, X., & Shi, R. (2024). Examining causal factors of traffic conflicts at intersections using vehicle trajectory data. International Journal of Transportation Science and Technology.
Zheng, Lai & Sayed, Tarek & Mannering, Fred. (2020). Modeling traffic conflicts for use in road safety analysis: A review of analytic methods and future directions. Analytic Methods in Accident Research. 29. 10.1016/j.amar.2020.100142.
王文麟(1996)，交通工程學理論與實用修正版。
本間正勝(Masakatsu Honma),宮田晋(Esusumu Miyata)(2002). 時差式信号制御の安全性向上に関する研究(Study on the Improvement of Safety at Signalized Intersections with a Lagging Green Phase).土木計画学研究・論文集 Vol.19 no.4.
交通部(2026)。道路交通標誌標線號誌設置規則。全國法規資料庫。https://law.moj.gov.tw/LawClass/LawAll.aspx?pcode=K0040014。
范玉琳(1979)。交叉路口綠燈早開或遲閉號誌控制之研究。國立成功大學土木工程研究所。
唐慧嚀(2006)。行車號誌倒數計時器設置程序之研究－以紅燈倒數計時器為例。﹝碩士論文。國立成功大學﹞臺灣博碩士論文知識加值系統。 https://hdl.handle.net/11296/zc53gw。
時差式信号機(2024年3月9日)。維基百科。https://ja.wikipedia.org/wiki/%E6%99%82%E5%B7%AE%E5%BC%8F%E4%BF%A1%E5%8F%B7%E6%A9%9F
高群凱(2022)。基於軌跡預測方法下之交通衝突指標探討： 以機會左轉為例。﹝碩士論文。國立陽明交通大學﹞臺灣博碩士論文知識加值系統。
張倩宜(2007)。號誌化路口衝突左轉停等車疏解行為之研究。中央警察大學。臺灣博碩士論文知識加值系統。
許添本、張家豪(1996)。號誌化交叉口之左轉肇事特性分析。八十五年道路交通安全與執法研討會。
許添本、温谷琳、張哲寧(2016)。應用完全貝氏法探討機車行駛第三車道肇事因子之研究。中華民國運輸學會。
許添本、溫谷琳、蔡牧融、楊皓宇、吳菘權、張開國、葉祖宏、賴靜慧、孔垂昌、黃明正(2021)。事故型態導向之路口交通工程設計範例之研究。交通部運輸研究所。
許添本、蕭唯倫(2018)。汽機車混合車流之車聯網防撞邏輯與參數之研究。《土木水利》 45卷2期 (2018/04) Pp. 18-31
黃文鑑、黃惠隆、林昶禛、鄭嵐瑜、施鈞明、陳一昌、賴靜慧(2010)。交通號誌時制重整計畫（III）－區域協調機制建立。交通部運輸研究所。
溫基信、王宏生、黃家耀、胡守任、蘇志文、張開國、葉祖宏、賴靜慧、孔垂昌(2021)。路口俯視攝影技術於交通衝突分析之案例應用與比較。交通部運輸研究所。
溫基信、王宏生、黃家耀、蘇志文、周業凱、楊植竣、張開國、葉祖宏、孔垂昌、黃明正(2020)。路口無人機交通攝影及衝突分析技術開發。交通部運輸研究所。
이은진(2023),교통신호기 설치·운영 업무편람(Traffic Signal Installation and Operation Guideline).경찰청(Korean National Police Agency),도로교통공단(Korea Road Traffic Authority), P81-89.