SEQUENTIAL UPDATE OF HIGHWAY TRAVEL-TIME FORECASTING USING A GREY MODEL

以灰方法建構具持續更新能力之高速公路旅行時間預測

李穎(Ying Lee) ； 魏健宏(Chien-Hung Wei) ； 徐賢斌(Hsien-Pin Hsu) ； 曾慶瑄(Ching-Hsuan Tzeng) ； 蘇郁涵(Yu-Han Su) ； 吳佩芸(Pei-Yun Wu)

《中國土木水利工程學刊》 31卷4期 (2019/06) Pp. 315-326

https://doi.org/10.6652/JoCICHE.201906_31(4).0003

摘要

Traffic conditions often change substantially over a short time. To decrease the uncertainty caused by the changing traffic conditions, this study applied speed data as the model input and demonstrated a univariant approach for travel-time forecasting models with sequentially updated traffic data by combining grey and regression methods. The rolling grey model (RGM(1,1)), incorporating prediction grey model (IPGM(1,1)), and incorporating partial prediction grey model (IPPGM(1,1)) were applied to forecast the speed by using historical speed data. Based on the forecasted speed from the grey models, the regression method was used to develop a functional relationship between the actual historical travel time from vehicle detectors and actual historical bus travel times. Consequently, the forecasted bus travel time was obtained by applying the forecasted travel-time from the grey models as the independent variable in the regression relationship. To reflect the actual traffic situations adequately, the data collection period included weekdays and weekends. For most links and paths, the mean absolute percentage errors (MAPEs) of forecasted bus travel times were lower than 9.6%, indicating a high-quality performance. For most links, the forecasted travel times computed from speeds forecasted by IPGM(1,1) and IPPGM(1,1) were more accurate than those forecasted by RGM(1,1). Empirical studies have shown that the proposed procedure effectively combines traffic data from many detectors to form travel-time information for travelers and traffic managers. The forecasted travel time can be calculated by inserting real-time traffic data into the function as required.

關鍵字

grey model ； speed forecast ； travel-time forecasting ； time-series ； tunnel area

並列摘要

為降低因交通變化對用路人造成的不確定感，本研究整合應用含預測訊息之灰預測與迴歸方法，從單變量時間序列角度開發具持續更新能力之旅行時間預測模式。本研究首先應用與開發滾動灰方法、含預測訊息之滾動灰方法與含部分預測訊息之滾動灰方法進行速率預測與旅行時間預測。研究分別使用平日與假日資料進行模式建構。路段與路徑旅行時間預測的研究結果具有高度準確率，其平均絕對誤差百分比皆低於9.6%。實證結果顯示，本研究所提出的資料處理程序與研究步驟可有效整合路段上各點交通資料，持續更新提供高準確的旅行時間預測訊息。

並列關鍵字

灰模型 ； 速率預測 ； 旅行時間預測 ； 時間序列 ； 隧道區

參考文獻

Dion, F.,Rakha, H.(2006).Estimating dynamic roadway travel times using automatic vehicle identification data for low sampling rates.Transportation Research Part B.40(9),745-766.

Park, D.,Rilett, L.(1998).Forecasting multiple-period freeway link travel times using modular neural networks.Transportation Research Record.1617,163-170.

Palacharla, P. V.,Nelson, P. C.(1999).Application of fuzzy logic and neural networks for dynamic travel time estimation.International Transactions in Operational Research.6(1),145-160.

Chien, I. J.,Ding, Y.,Wei, C. H.(2002).Dynamic bus arrival time prediction with artificial neural networks.Journal of Transportation Engineering.128(5),429-438.

Zhang, X.,Rice, J. A.(2003).Short-term travel time prediction.Transportation Research Part C.11C(3-4),187-210.