適當交通資訊之提供可作為高速公路用路人進行決策，尤其是針對與旅行時間有關之資訊，可做為駕駛者選擇適當之路徑與出發時間，避開擁擠車流，同時可選擇最短之旅行時間到達目的地，發揮高速公路之整體績效，再者，利用即時交通資料預測未來旅行時間，是為先進旅行者資訊系統中不可或缺之交通資訊。 本研究係針對國內高速公路施工路段進行研究，以微觀的角度探討施工路段之車流變化。在分析上主要分為兩階段進行，第一階段為利用實地調查之資料來界定施工路段之影響區，並分析施工路段內用路人變換車道之行為，進而撰寫模擬程式，作為交通資料之產生器。第二階段則是利用車輛偵測器所測得之交通資料，在各種不同流量、偵測器佈設間距與施工路段長度之實驗組合下，透過倒傳遞網路構建旅行時間預測模式，提供準確的預測，藉此作為用路人進行決策時判斷之依據。 經由反覆的校估與測試結果可以得知，本研究所構建之旅行時間預測模式，其預測效果良好，屬於「高精準預測」，於高速公路旅行時間相關資料提供方面，可做為相關單位參考之用途。 Provide the appropriate traffic information who can make a decision for highway drivers allow drivers to select appropriate routes and departure time to avoid congestion, especially deal with travel time. This traffic information can make drivers select the shortest travel time to arrive destiniation, and elaborate the whole effects for highway network, in addition, when using real time traffic information, it is necessary to forecast future travel time for Advanced Traveler Information System. This research is aimed at work zones in highway, studying the change of traffic flow with a microscopic view. There are two phases in the research. First, using data to define the effect range of work zones and analyze the change lane behaviors of drivers in live investigate. Take one step ahead, we write simulation formula which produce traffic information. Second, the research discuss different traffic flow, different location length of vehicle detector and different work zones length optained by vehicle detector. Utilizing backpropagation algorithm to build forecasting highway travel time models can provide exact predition. This result can be the base of making decisions for drivers. After repeatedly correcting and testing, the effect of forecasting model constructed in the research is very well, and belongs to high correct forecast. The research can be provided to forecast travel time for highway travel time estimation.
