Texas Highway Network Congestion Analysis

Keywords: highway traffic congestion, multi-criteria
Abstract Type: Paper Abstract

Abstract

In Texas, over 500 million hours of congestion occur annually within 10,000 miles highway network. Over 190 million gallons of fuel is wasted in congestion, and costs over $10 billion. However, there are challenges in defining the concepts of congestion, and the causing factors to congestions. In this study, we analyzed spatial distribution and temporal variation of the congestion concerned factors, including annual traffic volumes, accident statistics, and road segments network maps, etc. over the past 5 years in Texas highway inventory. By applying multi-criteria decision modeling algorithms (WSM, AHP, and TOPSIS), and aggregating criterions mentioned above among over 3600 highway segments in Texas, the scores interpreting the congestion situation were produced.
The post processing jobs in this project primarily focus on analyzing which factors resulting in the highway congestions in the top congested segments. Plus, by implementing sensitivity checking of multiple attributes, the high-order correlation of various pairs of causing factors were analyzed and visualized. Besides, we adopted and implemented a new method of analysis of congestion by calculating the alternative paths from one random starting point to another. In the future, this methodology could be applied to other state-wide traffic network not limited in highway congestion situation analysis, with a need of smaller size of datasets through inclusion of additional confounding variables into congestion index.