《Applying short-term traffic prediction models for updating missing values of traffic counts.doc》由会员分享，可在线阅读，更多相关《Applying short-term traffic prediction models for updating missing values of traffic counts.doc（45页珍藏版）》请在金锄头文库上搜索。
1、Applying Short-term Traffic Prediction Models for Updating Missing Values of Traffic CountsBy Ming Zhong,1 Pawan Lingras,2 and Satish Sharma3ABSTRACT: The presence of missing values is an important issue for traffic data programs. The analyses applied to data sets from two highway agencies show that
2、 significant portion of data has missing values. Literature review indicates that previous research mainly focused on detecting missing values. There is limited research on data imputation in traffic analysis. In this study, genetically designed neural network models and regression models were appli
3、ed to six permanent traffic counts (PTCs) from Alberta, Canada to investigate their merits in imputing missing values. These six PTCs belong to different trip-pattern groups and functional classes. A top-down model refinement was used to search for the models with reasonable accuracy for each type o
4、f road. Average errors for refined models were lower than 2% and the 95th percentile errors were below 4-5% for counts with stable patterns. Even for counts with relatively unstable patterns, average errors were below 6% in most cases, and the 95th percentile errors were rarely more than 10%. It is
5、believed that the models proposed in this study would be helpful for highway agencies in their traffic data programs. KEY WORDS: Missing Values, Short-term Traffic Prediction, Traffic Counts, Genetic Algorithms, Neural Network, Regression Analysis1Doctoral Student Faculty of Engineering, University
6、of Regina, Regina, SK, Canada, S4S 0A2 Phone: (902) 496-8152Email: Ming.Zhongstmarys.ca2Associate ProfessorDept. of Mathematics and Computing Science, Saint Marys University, Halifax, NS, Canada, B3H 3C3 Phone: (903) 420-5798Email: Pawan.Lingrasstmarys.ca3Professor Faculty of Engineering, University
7、 of Regina, Regina, SK, Canada, S4S 0A2Phone: (306) 585-4553Email: Satish.Sharmauregina.caINTRODUCTIONMissing values in traffic counts is an important issue for traffic engineers and data analysts, since traffic data programs were established in the 1930s (Albright 1991a). Dealing with missing value
8、s has significant implications to the limited budgets of traffic data programs. This was particularly true before the advanced data collection tools, such as mechanical detectors and electronic or ultrasonic sensors were applied in the fields, because of the high cost of manual counts (Sharma 1983).
9、 Even though scientists and engineers have managed to decrease the cost of data collection, highway agencies still commit a significant portion of their resources to data collection, summarization, and analysis (Sharma et al. 1996). Highway agencies need to use the collected data efficiently. This s
10、tudy analyzed missing values for the data sets from two highway agencies in North America. First data set was from Alberta Transportation Department and the other was from the Minnesota Department of Transportation. In Alberta, over seven years, more than half of total counts have missing values. As
11、 shown in Figure 1, during some years the percentage is as high as 70% to 90%. Minnesota data shows more than 40% counts having missing values. Williams et al. (Williams et al. 1998) also reported that approximately 20 percent of the data in the development and test sets of their study were missing.
12、 For the traffic counts with missing values, highway agencies usually either retake the counts or estimate the missing values. Estimating missing values is known as data imputation. Since sometimes retaking counts is impossible due to limited resource and time, imputing the data has become a popular
13、 method (Albright, 1991a). The experience with data from Alberta Transportation indicates that the agency used data imputation before 1995. The replaced values of missing data were marked with minus signs for some years. Imputing data with reasonable accuracy may help establish more cost-effective t
14、raffic data program. The analysis of Alberta data also shows that a significant percent (varied from 10% to 44% from year to year) of traffic counts have missing data for several successive days or months. Usually these PTCs can not be used to calculate AADT or DHV due to the missing data. Such PTCs
15、 may be used as seasonal traffic counts (STCs) or short-period traffic counts (SPTCs) or just discarded by highway agencies. However, the information contained in these PTCs is certainly more than that from STCs and SPTCs. If missing data from PTCs can be accurately updated, further analysis could b
16、e applied based on AADT or DHV. There are increasing concerns about data imputation and Base Data Integrity. The principle of Base Data Integrity is an important theme addressed in both American Society for Testing and Materials (ASTM) Standard Practice E1442, Highway Traffic Monitoring Standards (America 1991) and the American Association of State Highway and Transportation Officials (AASHTO) Guidelines for Traffic Data Programs (America 1992).
