在交通量迅速成長，道路日益擁擠的情形下，交通管理系統逐漸倍受重視，而系統中主要包含動態旅行資訊與交通控制，其中以動態交通指派 (DTA，Dynamic Traffic Assignment)模型為其核心，而起迄點 (OD，Original-Destination)需求量為動態交通指派(DTA，Dynamic Traffic Assignment)中不可或缺的輸入值，作為模擬及預測的基礎。除此之外，起迄點需求量還可應用在其他領域，如交控、意外管理及運輸規劃。因此，針對起迄點流量推估之研究已從一開始的靜態推估發展到即時性推估，以能應用在各種動態的模擬控制軟體。

在獲得起迄點需求量方法上，如進行流量調查是一項需要大量人力及時間的調查工作，因此建立一即時性起迄點流量推估模型除可降低困難度及節省經費外，還可符合交通的動態情形，準確預測未來起迄點間的流量，以提供交控或交通指派使用。因真實動態起迄點需求量資料難以獲得，因此本研究將利用交通模擬軟體所產生之路段流量資訊推估起迄點需求量，所採用之模擬軟體DynaTAIWAN (Dynamic Traffic Assignment and Information in Wide Area Network)為一區域級的核心交通分析與預測系統，主要是以國內交通特性為基礎所發展出的一套交通模擬軟體，研究中將藉由DynaTAIWAN進行模擬，取得路段流量資料，預測更新下一階段的起迄點需求量。

考量路網型態，本研究將採用模擬資料藉由DynaTAIWAN進行模擬，使用具有國內交通特性之交通模擬軟體，主要方法使用卡門濾波模式。希望能就目前國內交通特性，探討一般路網下與混合車流的情況下之OD推估情形，提供交控、意外管理及運輸規劃相關應用。

Intelligent Transportation Systems (ITS) aim to utilize the transportation system efficiently by strengthening the connection between traffic control measures and available information, such as real-time information and historical flow information. Traffic flow distributions are detected by surveillance systems, and the information is transmitted to the traffic management center. However, applications based on static OD flows do not capture the dynamics of build up and dissipation of congestion, time-dependent OD demands are extremely important in Dynamic Traffic Assignment, a core model in ITS to analyze dynamic flow distributions.

To generate O-D demand data through field surveys is a time consuming process. Dynamic O-D estimation can save human resources and reduce expense. Most of dynamic OD estimation methods are constructed based on traffic flow counts; however, the interrelationships between link traffic counts and OD are not clear.

In this research, a time-dependent O-D estimation algorithm, based on Ashok’s algorithm, is constructed under mixed traffic flow conditions. The framework is a Kalman Filter based approach, and the deviations of OD flows from historical estimates are used in an autoregressive process. The algorithm is a assignment-based model, in which Dynamic Traffic Assignment Models need to be incorporated within the algorithm. DynaTAIWAN, a simulation-assignment model, is used to generate time-dependent assignment results for dynamic OD estimation.

Numerical experiments and sensitivity analysis are conducted to illustrate the algorithm in a wide variety of scenarios.

•交通部 (1990)，「交通工程手冊」，幼獅出版社。
•李宗憶 (1998)，「考慮動態OD推估之適應性整體匝道儀控模式研究」，淡江大學運輸科學研究所碩士論文。
•卓訓榮、曾國雄、周幼珍、江勁毅 (1997)，「動態流量推估動態O-D方法之研究」，運輸計劃季刊，第二十六卷第四期，頁618-638。
•胡大瀛等 (2004a)，「區域級智慧型運輸系統示範計劃－核心交通分析與預測系統(第一年期)」，交通部運輸研究所。
•胡大瀛等 (2004b)，「區域級智慧型運輸系統示範計劃－核心交通分析與預測系統(第二年期)」，交通部運輸研究所。
•胡大瀛等 (2006)，「智慧型運輸系統下之核心交通分析與預測系統：即時控制模組開發(2/2)」，交通部運輸研究所。
•胡守任等 (2001)，「智慧型運輸系統基礎理論系列研究(一)－濾波理論(Filtering Theory)應用於流量倒推旅次起迄量(OD)及車流密度之推估」，交通部運輸研究所。
•凌瑞賢 (2001)，「運輸規劃原理與實務」，鼎漢工程顧問公司。
•陳齊邦 (2004)，「高速公路動態旅行時間與旅次起迄推估之研究」，淡江大學運輸管理學系運輸科學所碩士論文。
•張志浩 (2003)，「使用高斯狀態空間模型與旅行時間估計動態旅次起迄」，交通大學統計學研究所碩士論文。
•蘇怡如 (2005)，「供應鍊管理下需求預測控制之研究」，逄甲大學交通工程與管理學系碩士論文。
•Ashok, K. and Ben-Akiva, M. E. (1993), “Dynamic Origin-DestinationMatrix Estimation and Prediction of Real-Time Traffic ManagementSystems.”, Transportation and Traffic Theory, pp. 465.
•Ashok, K. and Ben-Akiva, M. E. (2000), “Alternative Approaches for Real-Time Estimation and Prediction of Time-Dependent Origin-Destination Flows.”, Transportation Science 2000 INFORMS, Vol.34, pp. 21-36.
•Ashok, K. (1996), “Estimation and Prediction of Time-dependent Original-Destination Flows.”, Ph.D. dissertation, Center of Transportation Studies, Massachusetts Institute of Technology, Cambridge, MA.
•Bell, M. G. H. (1991), “The Estimation of Origin-Destination Matrices by Constrained Generalised Least Squares.”, Transportation Research Part B, Vol. 25, pp.13-22.
•Cascetta, E., and Nguyen, S. (1988), ”A Unified framework for estimating or updating origin/destination matrices from traffic counts.”, Transportation Research B, Vol. 25, pp. 437-455
•Chang, G. L. and Tao, X. (1996), “Estimation of dynamic network O-D distribution.”, Transportation and Traffic Flow Theory, Elsevier Science.
•Chang, G. L. and Tao, X. (1999), “An Integrated Model for Estimating Time-varying Network Origin-Destination Distribution.”, Transportation Research Part A, Vol. 33, pp.381-399.
•Chang, G. L. and Wu, J. (1994), “Recursive Estimation of Time-varying O-D Flows from Traffic Counts in Freeway Corridors.”, Transportation Research B, Vol. 28, pp. 141-160.
•Chan, G. L. and Wu, J. (1996), “Estimation of time-varying origin-destination distributions with dynamic screenline flows.”, Transportation Research B, Vol. 34, pp. 277-290.
•Insightful, S-PLUS® 7—Delivering the Power of Predictive Analytics Across the Enterprise,http://www.insightful.com/products/splus/default.asp
•Nihan, N. L. and Davis, G. A. (1987), ”Recursive estimation of origin-destination matrices from input/output counts.”, Transportation Research B, Vol. 21, pp. 149-163.
•Nihah, N. L. and Davis, G. A. (1989), “Application of Prediction-Error Minimization and Maximum Likelihood to Estimate Intersection O-D Matrices from Traffic Counts.”, Transportation Science, Vol. 23, pp.77-90.
•Nihah, N. L. and Davis, G. A. (1991), “Stochastic process approach to the estimation of origin-destination parameters from time series of traffic counts.”, Transportation Reasearch Record 1328, pp. 36-42.
•Nihan, N. L. and Hamed, M. M. (1992), “Fixed-point approach to estimating freeway origin-destination matrices and the effect of erroneous data on estimate precision.”, Transportation Research Record 1357, pp. 18-28.
•Okutani, I. and Stephanedes, Y. J. (1984), “Dynamic prediction of traffic volume through Kalman filtering theory.”, Transportation Research Part B, Vol. 18, pp. 1-11.
•Spiess, H. (1987), “A maximum likehood model for estimating origin-destination matrices.”, Transportation Research Part B, Vol. 21, pp. 395-412.
•Van Der Zijpp, N. J. and De Romph, E. (1997), “A dynamic traffic forecasting application on the Amsterdam beltway.”, International Journal of Forecasting 13, pp. 87-103.