本所前於102年4月正式出版「自行車道系統規劃設計參考手冊」，106年配合「104年自行車環島串連路網標誌標線試辦計畫」法制化作業完成及相關法規條文修正進版，推出「自行車道系統規劃設計參考手冊（2017修訂版）」。近年相關法規有許多與自行車相關條文之修訂，加上人本環境之重視與自行車旅遊之推展，自行車道之配置與指示系統亦需檢討納入手冊修訂更新，以提供各單位規劃設計參考應用，爰檢討並提出「自行車道系統規劃設計參考手冊（2025年修訂版）」。修訂後本手冊內容共有七章，第一章總論說明本手冊編定之目的、手冊內容架構及手冊使用方式（適用範圍、對象及應用）。第二章說明自行車路網及路線規劃原則，藉由路網及路線之友善規劃，做為後續自行車道各項基礎設施設計之基礎。第三章說明自行車使用道路型式之基本分類、使用型式選擇流程及施工交通持等原則。第四章說明自行車道幾何設計原則，包括自行車安全視距、車道寬度要求、側向安全淨寬及淨高、自行車道線形、交叉路口穿越及立體穿越設施設計等規定。第五章說明自行車道鋪面暨附屬設施之設計原則，包括鋪面、排水、欄杆、分隔方式、自行車牽引道、編號及里程標示、導覽牌、自行車停車空間、自行車道照明、自行車道植栽及自行車補給站設置原則等。第六章說明自行車道標誌、標線及號誌等交通工程布設基本原則。第七章說明自行車道計畫評估與維護管理建議，包括完工自主檢視項目與標準、自行車道路線資訊屬性資料庫填報說明、巡查項目與頻率等。本手冊凡涉及現有規範內容者，均以粗體加底線方式呈現，並加註依循規範（依循規範可詳參附錄1-1），俾利規劃者能更精準掌握現有法規規定。The Institute of Transportation, Ministry of Transportation and Communications (MOTC), officially published the "Bikeway System Planning and Design Reference Manual" in April 2013. In 2017, following the completion of the legalization process for the "2015 Pilot Project on Signage and Markings for the National Cycling Route Network" and the amendment of related regulations, the “Bikeway System Planning and Design Reference Manual (2017)” was released.In recent years, numerous amendments to regulations related to bicycles have been made. Along with the growing emphasis on people-oriented environments and the promotion of bicycle tourism, the layout and signage systems of bikeways also require review and updates. These revisions will be incorporated into the manual to serve as a reference for planning and design by relevant agencies, leading to the release of the “Bikeway System Planning and Design Reference Manual (2025).”The revised manual consists of seven chapters. Chapter 1, General Introduction, outlines the purpose of the manual, its content structure, and instructions for use, including its scope of application, target users, and methods of application. Chapter 2 explains the principles of bikeway network and route planning. Through the user-friendly planning of networks and routes, it provides the foundation for the subsequent design of various bikeway infrastructure elements. Chapter 3 describes the basic classification of roadway types for bicycle use, the selection process for appropriate bikeway types, and principles for maintaining traffic flow during construction. Chapter 4 explains the geometric design principles of bikeways, including bicycle stopping sight distance, required lane width, lateral and vertical clearances, bikeway alignment, intersection crossing design, and grade-separated crossing facilities. Chapter 5 outlines the design principles for bikeway pavement and associated facilities, including pavement structure, drainage, guardrails, separation methods, bicycle ramps, route numbering and mileage markers, directional signage, bicycle parking spaces, bikeway lighting, landscaping, and the installation of bicycle service stations. Chapter 6 outlines the fundamental principles of traffic engineering design for bikeways, including signage, pavement markings, and traffic signals. Chapter 7 provides recommendations for bikeway project evaluation and maintenance management, including post-construction self-inspection items and standards, instructions for completing the bikeway route information attribute database, as well as inspection items and frequencies. Every regulation referred to related criteria was bolded and underlined for easy to read and use. Referred specifications were summarized in Appendix1-1.
為因應國內各項鐵道系統建設需求，交通部運輸研究所推動了一系列鐵道容量研究計畫。其中，在傳統鐵道方面，已研發鐵道容量分析模式與軟體，做為辨識容量瓶頸及評估運能提升的工具。然而，編組站與端末站容量分析的相關課題仍有深化空間。爰此，交通部運輸研究所啟動為期兩年的「城際鐵道容量分析暨應用研究」計畫，以持續精進相關研究。在第一年（民國112年）的研究中，已完成場站容量分析模式的開發。為進一步降低模式的使用門檻，第二年度（民國113年）研究重點著重於擴充「傳統暨區域鐵路系統容量分析軟體」，新增場站容量分析功能，以提升分析效率與實用性，而且特別針對自訂車站提供月臺與軌道配置編輯功能，讓使用者能更靈活地設計車站配置。此外，研究中考量列車受道岔速限影響及進出調車場等情境，研發出更精確的號誌安全時距計算公式，做為未來鐵道容量研究的重要基礎。最後，透過手冊編訂、成果發表會與教育訓練等方式推廣研究成果。To meet the needs of constructing various railway systems in Taiwan, the Institute of Transportation, MOTC has carried out a series of rail capacity research projects. In terms of conventional railway systems, rail capacity analysis models and software have been developed as assessment tools to identify capacity bottlenecks and evaluate capacity enhancement strategies. However, the capacity analysis issues at classification yards and terminal stations still need further investigations. Therefore, the Institute of Transportation, MOTC launched this two-year project titled “Research on Capacity Analysis of Intercity Railways and Its Applications” to advance and refine related studies.In the first year (2023) of this project, a station capacity analysis model was successfully developed. The second year (2024) focuses on enhancing "Conventional Rail Capacity Software" by integrating the station capacity analysis function to improve analysis efficiency and practicality. Notably, the software now includes features for editing platform and track layouts for custom stations, offering users greater flexibility in station design. Additionally, the study accounted for scenarios such as train speed restrictions at turnouts and entry/exit operations at classification yards, and developed more accurate formulas for calculating signal close-in time, which will serve as an important foundation for future rail capacity research. Finally, the project's outcomes have been promoted through manual compilation, results presentations, and training programs.
為提升機場空側營運效能及工程應變決策能力，本所與財團法人成大研究發展基金會合作研發本土化機場模擬技術，建置「機場空側模擬分析系統」( Airport Airside Simulation and Analysis System)。可提供桃園機場公司與交通部民航局等相關單位做為機場營運決策分析工具，強化機場空側管理及方案評估，並可進一步撙節相關成本及提升機場營運效能。本軟體系統功能包括機場空側場面管理、航班管理、專案管理及模擬評估等，使用者可自由設定機場場面配置、參數數值、機場起降航班數量、起降間隔，以及設定空側運轉情境，包括：起飛、降落、滑行等，以及拖機與地停等實務操作；並可因應不同天候及工程變動情境，進行方案評估與擇選。在開發過程中，已完成桃園機場多項實際案例分析及驗證，且透過機場相關機關(構)的試用，蒐集使用者回饋意見，並於113年10月在民航局與桃園機場公司各完成1場教育訓練，除上述單位人員參加以外，並邀請飛航服務總臺、臺北國際航空站、高雄航空站、臺中航空站等人員參與，以加強後續實務應用。以往我國機場因應營運所需進行之機場模擬分析，均委由國外團隊辦理專案，無法貼近營運使用需求(如場面例行維護之影響評估)，亦無法厚植我國機場空側模擬分析能力。本軟體以本所106~107年、110~111年之系統模擬技術為基礎，結合飛航與機場空側運作、等候理論、數學規劃排點與排程等模式，研發完成之機場空側容量評析核心技術，再進一步開發成為具人機介面親和性之完整軟體，可掌握機場空側容量、評估機場在日常情境及干擾情境下之空側配置方案優劣、釐清不同情境及場面施工所導致之瓶頸與延滯原因、節省航空公司經營成本及減少旅客等候時間，進而提升機場營運效能與競爭力。ABSTRACT:To improve airport airside operational effectiveness and decision-making capabilities in engineering responses, the Institute of Transportation, MOTC (henceforth the IOT) has collaborated with the NCKU Research and Development Foundation to research and develop the localized airport simulation technology, resulting in the creation of the Airport Airside Simulation and Analysis System (henceforth This Software). This Software can be used as an airport operation decision-making analysis tool by the Taoyuan International Airport Corporation, the Civil Aviation Administration, MOTC, and other relevant units, thereby improving airport airside management and plan assessment while also lowering related costs and increasing airport operation efficiency.This Software’s system functions include airport airside scenario management, flight management, project management, simulation assessments, and so on. Users can freely create airport airside scenario configurations, parameter values, number of airport takeoffs and landings, takeoff and landing intervals, and airport airside operational scenario settings, such as takeoff, landing, and gliding, as well as practical operations, such as traction and ground parking. Furthermore, in response to situations involving climate and engineering change, plan assessment and selection have been carried out. Several actual case analyses and verifications at Taoyuan International Airport were completed during the development process. Additionally, user feedback and comments were gathered through trials conducted by airport-related agencies (institutions) and as of October 2024, the Civil Aviation Administration, MOTC, and Taoyuan International Airport each completed one education training session. In addition to the personnel from the aforementioned units, personnel from the Air Navigation and Weather Services, Taipei International Airport, Kaohsiung International Airport, and Taichung International Airport were invited to participate in order to strengthen future practical applications.Previously, airport simulation analyses conducted in response to the operational needs of Taiwan’s airports were delegated to foreign teams as projects, which were not tailored to meet operational demands (such as the impact assessment of routine scene maintenance) or lay groundwork capabilities for Taiwan’s airport airside simulation and analysis capabilities. This Software is built on the foundation of IOT system simulation technology from 2017 to 2018, and 2021 to 2022. It, along with aviation and airport operations, waiting theory, mathematical planning and scheduling, and other models, has compelled the development of airport airside capacity assessment and core analysis technology. It has been developed into a comprehensive software with human-machine interface affinity intended to monitor airport airside capacity, the benefits and drawbacks of airport airside configuration plans in everyday situations, as well as interference situations. This helps to clarify bottlenecks and delays caused by various situations and scene configurations. As a result, airline operating costs can be reduced while passenger wait times are reduced, improving airport efficiency and competitiveness.
本計畫為二年期計畫，聚焦於「航港產業數位化調查與發展藍圖研擬」以及「研訂航港產業數位化發展指引」，以貨櫃運輸為研究對象，擘劃我國航港產業數位化發展方向，透過規劃數位化發展推動藍圖，擬訂相關推動策略。112年蒐集國內外航港數位化發展現況與趨勢，運用內容分析法歸納我國可借鏡之處，包括：數位與綠色雙軸轉型、港口社群系統、數位工具導入、網路資訊安全與港區安全管理、數據共享與資源共享；透過深度訪談法，歸納航港產業利益關係人之數位化需求、瓶頸與對政府建議；本研究亦蒐整我國官方現行運作之航港資訊系統內容架構，並建議從服務導向出發設計資訊整合架構，方便業者進行資料串接；最後根據研究成果提出發展藍圖雛型，並召開產官學研座談會，分別從公部門及私部門角度，規劃推動項目及進程，藉以廣泛蒐集意見，納入後續研究規劃參考。113年以前期研究成果為基礎並彙集政府機關與國營企業之意見，以公部門工作群組為平台，邀集航港相關公部門單位共同探討航運未來發展方向與策略，並促進跨部門交流與合作。藍圖依時間區分為短期(1-5年)、中期(5-7年)及長期(7-10年)三階段，共為政府部門及業者規劃32項策略，整體涵蓋系統面、資料面及產業面。在建立國內航港產業數位化評估指標架構方面，透過文獻解析法及層級分析法確定航港產業數位化的關鍵衡量指標，包括：數位轉型目標、新市場開發、資訊安全、數轉策略及創造價值等，並進行航港產業數位化問卷調查。本研究整合調查結果，最終提出涵蓋數位化價值、評估指標建置、現況分析、發展建議、案例研究與外部資源的《航港產業數位化發展指引》，以期協助業界規劃數位化策略並提供政策參考，推動我國航港產業加速數位化進程，提升競爭力與永續發展能力。ABSTRACT:Institute of Transportation initiated a two-year plan. This plan focuses on two key objectives: (1) Survey on digitization and Development Blueprint of Shipping Industry Ecosystem, and (2)Developing Guidelines for Digitalization of the Shipping Industry. This project takes container transportation as the research object, aiming to plan the digital development direction of the shipping industry in Taiwan through planning a digital development blueprint, and formulate relevant promotion strategies.The study in the first year emphasized collecting data and preparing a foundational development blueprint. began with an analysis of international and domestic digitalization trends, identifying actionable insights for Taiwan. In-depth interviews were conducted to identify stakeholders’ needs, bottlenecks, and suggestions. Additionally, the study reviewed the structure of the existing official shipping information systems in Taiwan and proposed a service-oriented approach to information integration. Based on these findings, a prototype blueprint was developed.The study in the second-year refined blueprint incorporated input from government agencies and state-owned enterprises through public-sector working groups. It outlined short-term (1-5 years), medium-term (5-7 years), and long-term (7-10 years) strategies, totaling 32 initiatives. These strategies span system infrastructure, data value-addition, port community systems, digital status improvement, talent cultivation, and data interface development.To establish an evaluation index for digitalization, the study used literature reviews and hierarchical analysis to identify critical metrics, such as digital transformation objectives, market development, information security, and value creation. A survey of the shipping industry's digital maturity was conducted.The study culminated in the development of the "Guidelines for the Digital Development of the Shipping Industry," which include: Digital value propositions, Evaluation indicators, Current situation analysis, Development recommendations, Case studies and external resources. These guidelines aim to assist the industry in formulating digital strategies and provide a reference for policymakers to accelerate the digital transformation of the shipping industry in Taiwan, thereby enhancing its competitiveness and supporting sustainable development.
鐵道系統為我國重要資產之一，其特性為高沉沒成本、完工後不易大幅調整路線，需要以長期及永續角度進行整體性規劃與維運，在面對氣候變遷極端氣候的壓力下，鐵路系統的挑戰不僅是要在正常天候中維持正常營運，更要在面對極端氣候的條件下增加其韌性，使其在受異常天氣衝擊後，能儘速恢復並提供安全、可靠且有效率的服務。本計畫於113年度主要回顧國外針對鐵道系統之氣候變遷調適文獻，包含發展趨勢、強化調適能力方式、新科技應用與相關實務案例等，同時盤點我國鐵道系統目前因應氣候變遷的調適方式，進而深入探討氣候變遷衝擊影響鐵道系統韌性強度之因素，並研擬鐵道系統因應氣候變遷強化調適能力的機制與方法。ABSTRACT:The railway system is a critical infrastructure in our country. It is characterized by high sunk costs and less flexibility in adjusting routes after completion. Thus, it requires comprehensive planning, operation, and maintenance from a long-term and sustainable perspective. Under the pressure of climate change, the challenges of the railway system are not only to maintain operation under normal conditions, but also to increase its resilience while facing extreme weather conditions so that it can recover as soon as possible to provide safe, reliable, and efficient services.This year’s (2024) research mainly reviews the international literature on climate change and adaptation for railway systems, including the development trends, the approaches to enhance adaptation capabilities, new technology applications, and related practical cases. It also reviews the current adaptation measures through which the railway systems in our country respond to climate change and investigates in-depth the factors affecting the resilience strength of railway systems. Finally, this research will propose the mechanism and methods to enhance the adaptation capability of railway systems in response to climate change.
本計畫聚焦於新興科技在學校交通安全教育中的應用，開發具擬真情境、互動性與遊戲化設計的教學輔助軟體，以提升學生交通安全意識與學習成效。本計畫依據交通部108年研訂之國小、國中、高中交通安全基本能力，以及111年新興科技導入交通安全教育之需求評估，針對行人、自行車與機車設計適用於不同學習階段的數位教材，並以平板裝置作為終端平台，開發沉浸式學習軟體。本計畫以行人、自行車、機車等主題設計，涵蓋第一人稱視角、第三人稱視角、情境問答與危險感知訓練，以強化反思學習。為提升課堂應用便利性，開發一對多情境體驗反饋後台，讓教師即時掌握學生學習成果。本計畫採分階段入校測試，於19所學校（國小6所、國中8所、高中5所）導入學習工具，涵蓋860名學生，並透過平板互動教學與前後測評估學生的知識成長與風險認知變化。結果顯示，學生測驗成績顯著提升，約86%的學生認為沉浸式學習有助於理解交通風險。教師回饋亦指出，數位學習工具能提高學生參與度，降低備課負擔，提升教學效率。本計畫證明，新興科技應用於交通安全教育能有效提升學習成效與風險意識，並為學校提供更具互動性與實務導向的教學模式。本計畫成果可供教育單位、政策制定者及交通安全推廣機構參考，以提升青少年交通安全素養，降低事故風險。ABSTRACT:This project focuses on the application of emerging technologies in school-based traffic safety education by developing an interactive, gamified instructional software with immersive scenarios to enhance students’ traffic safety awareness and learning effectiveness. Based on the Traffic Safety Core Competencies for Elementary, Junior High, and Senior High Schools established by the Ministry of Transportation & Communications in 2019 and the 2022 Needs Assessment for the Integration of Emerging Technologies into Traffic Safety Education, this project designs digital learning materials for pedestrians, cyclists, and motorcyclists at different educational stages. A tablet-based platform is used to facilitate immersive learning experiences.This project is designed around key themes such as pedestrians, bicycles, and motorcycles. It incorporates first-person and third-person perspectives, situational Q&A, and hazard perception training to enhance reflective learning. To improve classroom usability, a one-to-many scenario experience and feedback platform is developed, enabling teachers to monitor students' learning outcomes in real time.A two-phase school implementation was conducted across 19 schools (six elementary schools, eight junior high schools, and five senior high schools), engaging 860 students. The project utilized tablet-based interactive teaching combined with pre- and post-tests to assess students’ knowledge acquisition and risk perception changes. The results indicate a significant improvement in students’ test scores, with about 86% of students reporting that immersive learning helped them better understand traffic risks. Teachers' feedback also highlights that digital learning tools enhance student engagement, reduce lesson preparation time, and improve teaching efficiency.This project demonstrates that integrating emerging technologies into traffic safety education effectively enhances learning outcomes and risk awareness while providing schools with a more interactive and practice-oriented teaching approach. The outcomes of this project serve as a valuable reference for educational institutions, policymakers, and traffic safety advocacy organizations in promoting youth traffic safety awareness and reducing accident risks.
本計畫研究對象為梁式橋、板橋、箱型橋等三種類型的混凝土橋梁，研究目標為利用無人機拍攝橋梁構件影像，再透過AI深度學習語意分割技術，偵測影像上橋梁表面各種劣化類型，包括混凝土裂縫、混凝土剝落、鋼筋銹蝕、滲水、白華損傷等。1. 本計畫以兩種不同形式的橋梁當示範區，包括箱型梁與PCI梁， Y6B1200無人機在自動導航定位的定位誤差，在不使用UWB情況下，絕對定位精確度仍達30~50 公分，可減少人工操作之精神壓力，且可提供高解析影像之品質與穩定度。整體而言，箱型梁橋之高解析影像覆蓋率可達90%，但I型梁橋則受限於主梁間空間太小，高解析影像覆蓋率僅能達到60%。未來若要再提升覆蓋率僅能使用小型無人機，飛入主梁間狹小空間以更近的距離拍照。2. 本計畫建立一個基於深度學習技術的無人機影像品質評估模型，並製作無人機影像訓練資料集，包括亮度失真、對比度失真、高斯模糊、水平移動模糊和垂直移動模糊等退化影像，所獲得的影像品質評估模型能正確的篩選出品質及格(SSIM≥0.7)和品質不及格(SSIM3. 本年度針對112年所開發之Deeplab V3++橋梁影像劣化辨識模型進行精進，增加了2%高解析影像之標註，用於各劣化模型的微調訓練，使得新模型整體辨識率有提升，F1分數在裂縫、銹蝕、剝落、白華、滲水等劣化類別分別從39.35、25.86、54.33、34.86、11.21提升至53.70、75.41、66.46、69.72、60.85。4. 在幾何變形偵測部分，本計畫提出兩種方式，針對橋台、橋墩、基礎這類的構件，可以使用不同時期的3D點雲，透過點雲距離的計算分析變形量。而若是主梁或帽梁之類的構件，若有垂直面向下滑動的變形，則可以製作不同時期的正射影像，並透過特徵點匹配技術計算位移量或撓度，用以分析其變形量與變形型態。ABSTRACT:The project of this study are three types of concrete bridges: beam bridges, slab bridges, and box girder bridges. The research aims to use UAV to capture images of bridge components. AI deep learning-based semantic segmentation techniques are then applied to detect various types of surface deterioration on the bridge, such as concrete cracks, spalling, rebar corrosion, infiltration, and efflorescence.1. This project uses two different types of bridges as demonstration areas, including a box girder and PCI girder. The positioning error of the Y6B1200 UAV in automatic navigation, without using UWB, still achieves an absolute positioning accuracy of 30–50 cm. This reduces the mental stress of manual operation while providing high-resolution image quality and stability.2. This project established a UAV image quality assessment model based on deep learning techniques and created a training dataset of UAV images. The dataset includes degraded images such as brightness, contrast, Gaussian blur, horizontal motion blur, and vertical motion blur. The resulting image quality assessment model can accurately classify images as either passing (SSIM ≥ 0.7) or failing (SSIM 3. This year, improvements were made to the Deeplab V3++ bridge image defect recognition model developed in 2023. An additional 2% of high-resolution images were annotated and used for fine-tuning the AI model, resulting in an overall improvement in recognition accuracy. The F1 scores for different degradation categories improved as follows: cracks (39.35 → 53.70), rust (25.86 → 75.41), spalling (54.33 → 66.46), efflorescence (34.86 → 69.72) , and infiltration (11.21 → 60.85).4. Regarding geometric deformation detection, this study proposes two methods. For components such as abutments, piers, and foundations, 3D point clouds from different periods can be used to calculate and analyze deformation amounts by measuring point cloud distances. For components like main girders or cap beams, if vertical sliding deformations occur, orthoimages from different periods can be generated, and feature point matching techniques can be applied to calculate displacement or deflection, allowing for an analysis of deformation magnitude and patterns.
PortCDM為透過傳遞標準化資訊與即時共享資訊，有效提升船舶進出港過程之協作效率，減少等待時間及能源消耗，並進一步提升港口作業的可預測性與可靠性。本研究旨在評估我國建立PortCDM之可行性，探討如何透過提升資訊透明度與能見度，改善港口資源的使用效率，並藉由協助船舶調整航行速度來減少燃料消耗及碳排放。首先透過文獻回顧，分析國外港口實施PortCDM之案例，了解其架構、推行方式以及影響其執行因素，提供臺灣發展PortCDM借鑑。本研究探討基隆港、臺中港及高雄港，藉由與各相關利害關係人之訪談，分析國內資訊共享現況以了解國內當前面臨之困難及相關利害關係人於資訊傳遞方面之需求，並且繪製我國港口資訊傳遞圖，標示船舶進出港之資訊傳遞狀態與重要資訊節點。為評估測試PortCDM之可行性，根據國內港口資訊傳遞情況設計PortCDM案例模擬推播情境及相關問卷，邀請港口相關利害關係人參與測試，主動推播利害關係人所需之船舶即時進出港資訊至其通訊群組並透過問卷調查其意見。結果顯示，相關利害關係人認為測試模擬PortCDM在提升資訊透明度、提高預估資訊準確性和改善工作流程等方面具有效果，顯示國內具有發展PortCDM之可行性。最後對我國發展PortCDM之推行策略提出建議，包含主導者與經費來源等並提出短、中、長期之發展期程以作為我國未來推動PortCDM之參考依據。ABSTRACT:PortCDM enhances the efficiency of collaboration during vessel port calls through standardized information exchange and real-time data sharing. It effectively reduces waiting times and energy consumption while improving the predictability and reliability of port operations. This study aims to assess the feasibility of establishing PortCDM in Taiwan, exploring how increased information transparency and visibility can optimize the utilization of port resources and assist vessels in adjusting sailing speeds to reduce fuel consumption and carbon emissions.This study begins with a literature review, analyzing international cases of PortCDM implementation to understand their frameworks, implementation methods, and success factors, providing valuable insights for Taiwan's development of PortCDM. This study then focuses on Keelung Port, Taichung Port, and Kaohsiung Port, conducting interviews with stakeholders to analyze the current state of information sharing, identify the challenges faced, and investigate stakeholders' needs for information transmission. A Metro Map for Taiwan’s ports was developed, clearly illustrating the state of information transmission during vessel arrivals and departures. Additionally, key nodes and decision points were identified, serving as a foundational reference for the preliminary promotion of PortCDM in Taiwan.To further validate the effectiveness of PortCDM, a simulation case was designed based on current domestic challenges. Stakeholders from Kaohsiung Port participated in the test, where dynamic information was delivered to them in a "real-time" and "proactive" manner. A questionnaire was used to evaluate participants' experiences and the effectiveness of the simulation. The results indicate that stakeholders have strong demands for such a system and recognize the significant benefits of PortCDM in enhancing information transparency, improving predictive accuracy, and optimizing workflows. This indicates a high feasibility for the development of PortCDM in Taiwan.Finally, based on the findings from the interviews and simulation, this study proposes specific recommendations for Taiwan's future PortCDM implementation strategies, including leadership roles and funding sources. Furthermore, it outlines short-term, medium-term, and long-term development plans, providing a blueprint for promoting PortCDM in Taiwan and advancing smart port operations.
近年來政府對於道路安全的推動和改善不遺餘力，交通事故帶來的社會與經濟損失也是最大的問題，為改善道路交通安全，各級道安主管機關每年度從工程、教育、執法、監理、宣導等各面向，投入預算與人力執行各項道安改善工作，但由於道安改善工作缺少系統性整合、道安改善專業人才培育亦缺乏完整機制，以至於產生未有足夠專業人力得以支援道安改善工作的困境。本計畫針對道安知識平台架構、事故肇因分析、道安人員專業養成及道安專業輔導/諮詢團隊機制等面向進行系統性探討並研擬相關建議。期望透過建立道路安全專業生態系，整合道安專業量能，協助中央及地方有效進行道安改善作業，並透過建立專業培訓機制，提升從業人員專業素養，促進道路交通安全。In recent years, the government has been relentless in promoting and improving road traffic safety. The social and economic losses caused by traffic accidents are the biggest issues. To improve road traffic safety, various levels of road safety authorities invest budgets and manpower annually in various aspects such as engineering, education, law enforcement, supervision, and advocacy. However, due to the lack of systematic integration in road safety improvement works and the incomplete mechanism for professional personnel training, there is a dilemma of professional practitioners to support road safety improvement.This project systematically explores and proposes relevant recommendations on aspects such as the road safety knowledge platform architecture, accident analysis, road safety professional training, and professional road safety counseling/consulting teams. By establishing an ecosystem and integrating professional capabilities, assist central and local governments in improving road safety. Additionally, through the establishment of a professional training mechanism, enhance the expertise of practitioners and promote road traffic safety.
我國汽車運輸業長期面臨交通事故風險高、駕駛管理不易等問題，對道路安全與社會資源造成嚴重衝擊。如何善用新興科技提升營運安全，尤其在駕駛行為風險辨識與預測方面建立有效工具，是政府與運輸業者亟欲解決的關鍵議題。本四年期研究計畫「應用人工智慧分析技術探勘高風險路段」，以人工智慧與車載資料為核心，分階段開發駕駛行為分析工具與風險預警模型，並針對國道客運業者之實務需求進行系統驗證與推廣策略評估。前3年期計畫已完成異常事件的定義與辨識模組建構，整合ADAS警示事件、車內外影像、駕駛行為數據、車道幾何與環境因子等，建立駕駛人風險行為樣態資料庫。第四年期計畫的研究重點主要集中在三大面向：（一）整合風險要素建構駕駛行為預測模型；（二）識別高風險路段與匝道之空間分析；（三）將異常事件分析工具轉化為實務可行之管理系統。首先，在駕駛行為預測方面，計畫採用Boosted Regression Tree與SHAP值分析，依據「本車行為」、「環境互動」、「駕駛員特性與動作」三要素構建多變數風險模型。結果顯示，大多數模型之AUC值達0.8以上，具高度準確性；特定風險如車道偏移、變換車道，與駕駛年資不足、距離前車過近、車道曲率大等情境高度相關，揭示異常事件背後的交互機制。本期導入時間序列資料處理技術與深度學習模型，針對逐秒ADAS警示資料進行預測。測試結果指出，當自變數時間窗設為前10秒、應變數為後5秒時，可有效預測ADAS警示事件發生，且能於駕駛出現高風險動作前即時警示，強化業者事前預防管理能力。此外，SHAP值分析顯示，影響警示事件的重要變數包括車速變異度、方向燈使用、天候狀況與過往警示紀錄，顯示駕駛操作行為與外部環境密切交織影響風險結果。在空間分析方面，計畫針對「巨觀路段」與「巨觀趟次」兩種模式進行建模。巨觀路段分析以2公里為單位，運用負二項迴歸模式建立高風險路段識別工具，反映時空變異下之交通風險分布。巨觀趟次則以15公里為分析單元，建構多層次混合效果模型，處理駕駛人異質性對風險的影響，並導入車流量、班表與駕駛資歷等因素。這些模型不僅提供業者實用的行車前風險提醒，也能協助主管機關掌握潛在高風險區段，做為設置警示標誌與工程改善依據。影像辨識技術部分，本期評估影像辨識輕量化策略，針對車內影像偵測建議採間隔幀取樣以降低系統負荷，惟針對高速國道場景之車外影像辨識仍需維持高幀率，以確保事件連續性與準確度。同時，本期計畫亦納入第二家業者之資料進行系統轉移性評估，發現GPS紀錄頻率與DMS攝影角度為影響資料兼容性的關鍵變數。建議業者若欲採用本系統，應考量升級車機硬體與增設車內影像設備，以提升風險分析品質。為促進技術落地，本期計畫彙整業者、設備商與主管機關意見，提出概念驗證（PoC）、服務驗證（PoS）、商業驗證（PoB）三階段推廣架構，以逐步實現AI風險分析系統的商品化。考量本系統未來在市區道路環境的應用可能性，計畫亦初步評估其可行性，研究結果指出雖核心風險建模邏輯具轉移潛力，但影像辨識模組須因應市區道路遮蔽、車道不明與車種多樣等特性進行調整與再訓練。最終，本期開發完成的駕駛風險管理系統，具備異常事件警示儀表板、駕駛風險綜合報表、趟次風險分數計算與個人行為趨勢追蹤等功能，可協助客運業者即時發現高風險駕駛與路段，並進行針對性管理。此系統除可顯著降低資料處理成本，也提供主管機關一套具科學依據之空間風險分析工具。ABSTRACT:Taiwan’s commercial motor carrier sector has long struggled with a high incidence of road crashes, leading to serious casualties and substantial social costs. In response, this four-year research project titled “Applying Artificial Intelligence Techniques to Identify High-Risk Road Segments” aimed to develop an integrated system for analyzing driving behavior and possible risks, using in-vehicle data and AI technology. The project was executed in phases, progressively enhancing behavior analysis tools and validating them in collaboration with intercity bus operators. The focus of the fourth year was to consolidate data modeling, identify spatial risk patterns, optimize prediction tools, and translate analytical findings into practical safety management solutions.In the first three years, the project established a foundation for risk analysis by developing a set of abnormal event definitions, collecting large-scale telematics and video data, and building the initial behavioral risk indicators. The fourth-year research was structured around three major objectives: (1) integrating multidimensional factors to construct predictive models of risky driving behavior, (2) conducting spatial analysis to identify high-risk road segments and ramps, and (3) transforming analytical outputs into an operational management system. Using Boosted Regression Trees and SHAP, the research team modeled risk based on three domains: vehicle behavior, environmental interaction, and driver characteristics and in-cabin actions. The models demonstrated strong predictive performance, with AUC values consistently reaching 0.8 or higher. Specific behaviors—such as lane deviation or unsafe lane changes—were found to be highly correlated with contextual factors such as insufficient experience, short headways, sharp curves, and mixed traffic conditions.To enable proactive intervention, the team implemented time-series modeling using deep learning algorithms (LSTM, RNN, 1D CNN-LSTM) to predict ADAS warning events in real time. With the optimal configuration—predictor windows of 10 seconds and a response window of 5 seconds—the models successfully forecasted warning events, allowing for risk alerts a few seconds before risky behaviors occurred. SHAP analysis further identified influential variables, such as speed variability, use of turn signals, urban context, previous alerts, and weather, thereby informing targeted interventions and driver feedback strategies.On the spatial analytics front, the project employed two complementary models: a “macro trip model” based on 15-kilometer driving segments and a “macro road segment model” using 2-kilometer units. The trip model incorporated multilevel mixed-effects regression to account for driver heterogeneity, while the segment model applied panel data negative binomial regression to capture monthly variations in traffic risk. These models produced cumulative risk scores, enabling fleet operators to issue pre-trip safety warnings and allowing government agencies to identify segments in need of engineering improvement or focused enforcement.In terms of image recognition technology implementation, the team assessed lightweight strategies for video processing. For in-cabin gesture detection (e.g., steering behaviors), frame sampling effectively reduced computational load without sacrificing performance. However, in highway environments requiring detailed motion detection, full-frame processing remained necessary to preserve temporal continuity. To test system portability, the fourth-year project also conducted a transferability analysis using data from a second operator. Results indicated that GPS logging frequency and DMS (Driver Monitoring System) camera angle were key determinants of compatibility. For full system functionality, operators were advised to upgrade hardware and ensure high-resolution, forward-facing driver views.To support future deployment, stakeholder engagement sessions were conducted with bus operators, equipment suppliers, and regulatory agencies. Based on feedback, a three-stage implementation roadmap was proposed—Proof of Concept (PoC), Proof of Service (PoS), and Proof of Business (PoB). The current system, having reached the PoC stage, includes driver-level dashboards, real-time alerts, and behavior trend reports. Field testing in cooperation with operators is expected to follow, with the PoS phase focused on embedding the system into daily fleet operations. The fourth-year project also considered the applicability of the system to urban driving contexts. While the conceptual framework—including event definition and risk scoring—remains valid, technical adaptations are required. Lane detection and spatial grids need refinement to handle occlusion and high-density mixed traffic, particularly involving smaller vehicles such as motorcycles. Therefore, technology transfer to urban settings will require retraining image recognition models and revising detection logic to account for urban roadway complexity.Finally, the project culminated in the development of a functional Driver Risk Management System, equipped with modules for visualizing abnormal events, tracking warning trends, scoring individual driver behavior, and issuing pre-trip alerts. This system helps fleet operators quickly identify high-risk drivers and road segments, significantly reducing the cost and effort required for data processing and analysis. For regulatory bodies, it provides a scientifically grounded platform for spatial risk profiling and targeted safety interventions.
「道路安全檢核Road Safety Audit -RSA」是指在道路建設專案規劃設計及通車前後，檢視其設計文件，針對安全議題，由獨立的檢核師所執行的一個正式的安全檢核程序。一般包括可行性研究、基本設計、細部設計到通車前的檢核，亦有針對現有道路進行檢核；有些國家將現有道路的安全檢查，區分開來稱為「道路安全檢查Road Safety Inspection -RSI」。此一道路安全檢查(RSI)是，針對既有道路，進行的一種系統性的現場檢查作業；由道路安全專家執行，辨識出可能導致嚴重事故的道路設施或道路安全缺陷。兩者同樣強調須由獨立檢查員或檢核員為之，並有一套完整詳細的程序，其目的主要是發現新建道路與既有道路的可能安全缺失、設施損壞，並及時加以更正，以避免導致事故發生，或降低事故發生的嚴重性。我國依據新修訂之公路法及新訂之道路交通安全基本法，皆訂有針對公路及道路進行安全檢核的要求。故而，本計畫旨在依據我國道路環境特性，參考各國道路安全檢核的發展經驗，研提我國道路安全檢核制度發展架構；同時基於此道路安全檢核制度發展架構，參酌國外對於現有道路安全檢查表內容，參考國內的道路環境及交通事故特性，研擬我國針對現有道路進行道路安全檢查之道路安全檢查表。其成果可供交通部、所屬交通管理單位，與各地方政府交通管理單位等參考，以便運用於既有道路的道路工程與交通工程的檢查與改善工作，來創造安全的交通環境。A Road Safety Audit (RSA) is defined as a formal procedure conducted during the project plan and design stage before any construction has started, screening the designs on paper for any safety issues. This is a formal process best conducted by an independent auditor. Conversely, a Road Safety Inspection (RSI) is a distinct review process for exiting roads, involving a systematic on-site assessment through driving and walking to identify hazardous conditions, defects, and deficiencies that may lead to serious accidents. Both RSA and RSI emphasize the necessity of independent inspectors or auditors and adhere to comprehensive procedure, with the primary objective of identifying and rectifying potential safety deficiencies or infrastructure damages in both new and existing roads to mitigate accidents or reduce the severity causing by them.In accordance with emendation to the Highway Law and newly promulgated Road Traffic Safety Law in Taiwan, the establishment of RSA and RSI systems is mandated. This project aims to develop a framework for the developmentation of the Road Safety Audit (RSA) system tailored to the specific characteristics of Taiwan’s road environment drawing upon the developmetal experiences of road safety audits in other nations. Furthermore, based on this framework and informed by road safety inspection checklists from international sources, a customized Road Safety Inspection (RSI) checklist will be developed for Taiwan. These deliverables will serve as resources for the Ministry of Transportation and Communications, its affiliated traffic management agencies, and local government traffic management units, facilitating the inspection and enhancement of road and traffic engineering on existing roads to ensure a seamless and secure traffic environment for all road users.
因應全球5G資通訊技術的快速發展，以及數位轉型與數位雙生等數位治理觀念，交通部運輸研究所(以下簡稱運研所)於民國110年提出「構建5G智慧交通數位神經中樞」的研究構想，並於110至111年間完成前期計畫，內容包括「構建5G智慧交通數位神經中樞(1/2)－功能架構探討與系統規劃」及「構建5G智慧交通數位神經中樞(2/2)－系統雛型開發與驗證實作」，初步完成5G智慧交通數位神經中樞的雛型系統開發，為後續發展奠定基礎。自112年起，運研所持續辦理112至113年度「構建5G智慧交通數位神經中樞－功能擴充與精進
本研究規劃我國都會地區、偏鄉地區及觀光旅遊地區智慧公共運輸服務之短、中、長期發展策略及行動方案，並研議協助地方政府及公車業者推動智慧公共運輸服務發展之補助項目、範圍及經費額度，同時研擬適合新一期公路公共運輸計畫之績效指標項目與分年績效值。此外，亦推估智慧公路公共運輸服務關聯產業產值，並評估新一期公路公共運輸計畫資金投入之效益值。研究成果可作為交通部相關單位於112年撰寫新一期公路公共運輸計畫之參據。
修訂淨零排放評估模型模型；更新運具碳排資料；提出第三期運輸部門溫室氣體排放基線；辦理淨零關鍵戰略管考與社會溝通會議；蒐集國際運輸場站節能減碳標竿案例；蒐研綠運輸生活型態體驗案例；彙編111年運輸部門溫室氣體減量行動方案成果報告。
本計畫探討偏鄉交通行動服務MaaS（簡稱偏鄉MaaS）之服務範疇、發展課題與推動策略，透過參考國際（如日本、芬蘭）以MaaS解決偏鄉交通議題之經驗，以及整合我國過去都會型MaaS推行成果與偏鄉運輸服務實績（如幸福巴士、噗噗共乘），探討以公共運輸為主體、連結在地觀光、地方創生、異業合作之機會，並就偏鄉區域特性提出偏鄉MaaS分類、服務模式、發展策略及可能之試辦場域規劃，以利建立可行之推動模式，期藉由交通服務智慧化、資源整合與人本導向設計，提升偏鄉交通可及性與使用者便利性。相關研究成果可做為未來推動拓展國內交
本計畫為期兩年，本期蒐集國內外共享運具管理規範和連結公共運輸的相關文獻及案例，分析共享運具於運輸系統之定位及服務模式。同時彙整共享運具整合公共運輸服務之相關規範及智慧化營運議題，訪談營運業者及政府部門，並針對我國共享運具暨連結公共運輸之發展定位、營運管理、服務整合、資訊應用等面向及第二年示範場域建置推動計畫書、不同共享運具連結公共運輸之指引等項目凝聚初步共識，據以提出具體策略及發展方向、國內共享運具與公共運輸智慧整合、創新加值及相關法規與規範之建議。
本計畫透過國際法制政策及綠色經濟、永續金融法制政策下的策略分析，針對標竿國家與我國現況進行落差分析；其次，透過汽車貨運業者及公私部門利害關係人訪談，盤點釐清企業困境及需求，提出區分為四大規範機制、管理規劃、設備技術、財政工具策略架構框架，再配合根據經濟、產業與技術發展等外在環境趨勢並針對我國汽車貨運業特性，選擇合適的淨零轉型政策並嘗試分析各項策略所涉及的課題，例如可能面臨之推動障礙，以作為未來行動方案規劃參考，協助我國貨運業者順利轉型，提升產業競爭力。
本計畫持續盤點人工智慧強化學習在號誌控制相關應用，在「都會區幹道強化學習號誌控制」部分，透過導入多任務強化學習演算法，以及通用環境之探討與建置，規劃以實測探討各時段與車流情境之泛用性，期望能提升人工智慧號誌控制模式之運作效率。在「交流道區域強化學習號誌協控」部分，以集中訓練/分散執行(CTDE)機制訓練多代理人，同時協調控制平面道路號誌與匝道儀控率，完成高速公路交流道匝道儀控與平面道路路口號誌協控之人工智慧強化學習號誌控制模型的發展與實作，及其運作量化效益分析，實測績效顯示：(1)臺北市中「中山北路與德行
本研究回顧臺灣地區易肇事路段改善計畫的歷史沿革，並分析其發展趨勢與面臨的挑戰。研究發現計畫自69年起推動至今，已進行41期，對降低交通事故發生率與死傷人數具顯著成效。然而，自92年起改善經費改由縣市政府自籌，導致提報改善地點及資金投入逐年下，產出之報告流於彙整少數提報地點，面臨全臺灣易肇事路段改善績效代表性不足之問題，此外地方政府在道安人力與資源配置方面仍有提升空間，且適逢行政院核定「國家道路交通安全綱要計畫(113至116年)」後，原依據之「第14期院頒『道路交通秩序與交通安全改進方案』(112年-11
本研究辦理「汽機車混流衝突」、「左轉車道配置與行車動線」兩項先導測試計畫，提出三大類駕駛行為：具風險之駕駛及用路行為、違規行為、交通衝突，並研提相關軌跡判斷邏輯，完成風險駕駛行為或違規行為的偵測，亦持續優化相關分析技術，可提供道路主管機關及交通管理者分析道路交通衝突，以提出更適切的改善方案。
