United States Department of Transportation
ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners.
As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.
i
AI-Enabled Transportation Network Analysis, Planning and Operations
-
2023-08-31
-
[PDF-14.31 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
DOI:
-
Resource Type:
-
Geographical Coverage:
-
Edition:Final Report (January 2022 – August 2023)
-
Corporate Publisher:
-
Abstract:This study introduces a unified end-to-end framework for analyzing network traffic equilibrium. The framework learns supply and demand components directly from traffic data, using computational graphs to parameterize unknown elements. It enforces user equilibrium through variational inequalities and can incorporate various modeling approaches, including neural networks. A novel neural network architecture is proposed that guarantees equilibrium states and allows for future scenario planning. The model is trained using advanced gradient descent algorithms and leverages operator-splitting methods for solving variational inequality problems. The framework's effectiveness is confirmed through tests on three synthesized datasets.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
