01137514

Component

Average annual daily traffic (AADT) values play a key role in many models and policy decision making processes in transportation and infrastructure management. Current widely used methods, however, only give rough estimates and the results are far from satisfactory. In this paper, the most commonly used approach, which estimates AADT by applying appropriate seasonal and day-of-week coefficients to the short term, is examined and other methodology that intends to modify this approach is reviewed. Based on the analysis of the problem and the limitation of the existing estimation approach, a framework for estimating AADT using co-clustering based collaborative filtering (CCCF) is proposed. Also, the details of implementation of the proposed framework, including the static and dynamic processes of learning from both recorded data and updated data, and a fast well-behaved co-clustering algorithm, are included in the paper. The results in the case study show that the proposed framework not only provided us insight to the data collection configuration, but also gave much better results compared to the existing factoring approach in terms of accuracy with low computation costs.

01120148

09-3432

English

16p

2009

Transportation Research Board 88th Annual Meeting

DVD

Figures (5) ; References (14) ; Tables (3)

Algorithms; Annual average daily traffic; Error analysis; Estimating; Traffic estimation; Traffic models; Traffic volume

Data and Information Technology; Highways; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2009 Paper #09-3432

BTRIS, TRIS, TRB

Jan 30 2009 7:50PM