.png){kind=icon}
Return
Classifying road network patterns using multinomial logit model
.png){kind=icon}
OA
AI
DOI：10.1016/j.jtrangeo.2016.11.013.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Road network patterns influence traffic operational performance and road safety. A precise description of road network patterns can provide useful guidance for both design and improvement of road systems. Most previous studies have classified road network patterns using a visual inspection method. Although visual inspection is valuable in practical applications, it can be subjective and time consuming. Hence, a more reliable, automated, and quantitative approach for the classification of road network patterns is required. The objective of this study is to develop and apply a quantitative method for network pattern classification. Data analyzed are from 718 traffic analysis zones (TAZs) in Florida's Hillsborough County. Six quantitative metrics, geometric and topologic, were analyzed, and their ability to classify different road network patterns were compared. Then, a multinomial logit model was developed to classify various network patterns using the six metrics. The results show that meshedness coefficient, proportion of cul-de-sacs, and proportion of 4-legged intersections were the three most significant variables in determining network patterns. Finally, this quantitative method was validated using TAZ data from Florida's Orange County, with an accuracy of 74.7%. This accuracy demonstrates the potential of the method in automatically and reliably classifying road network patterns. (C) 2016 Elsevier Ltd. All rights reserved.
Keywords:
Road network patterns
Traffic analysis zone
Geometric metrics
Topologic metrics
Multinomial logit model
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.3
Papers:
3.8K
Citations:
1.7W
Organization
Cited Papers
ShareSaveAn Efficient Algorithm for Nonlinear Model Predictive Control of Large-Scale Systems Part I: Description of the Method (Ein effizienter Algorithmus für die nichtlineare prädiktive Regelung großer Systeme Teil I: Methodenbeschreibung).png)
auto
autoIF0
PREAI
ShareSaveShareSaveShareSaveShareSave
View more