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
Management of Supply Chain Disruption of Freight Network Using Advanced Algorithms
-
2022-08-01
[PDF-1.65 MB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:The growing complexity of global supply chains and the rising frequency of transportation disruptions have underscored the need for predictive tools that enhance freight network reliability. This study explores the integration of big data analytics and machine learning to support short-term travel time prediction and incident detection in freight transportation systems. Using the National Performance Management Research Data Set (NPMRDS) for the state of Florida, a comprehensive dataset was developed, incorporating temporal, spatial, and traffic condition features at 5-minute resolution intervals over a one-year period. Four machine learning models, XGBoost, Random Forest, Decision Tree, and Multilayer Perceptron (MLP), were evaluated for travel time regression and binary incident classification. The results demonstrate that tree-based ensemble models (particularly XGBoost and Random Forest) achieved superior predictive accuracy, with high R² scores for travel time prediction and high precision-recall performance for incident classification. Feature engineering techniques, including cyclical time encoding, significantly enhanced model performance. The findings confirm the value of leveraging open transportation datasets and machine learning for improving operational awareness and supporting resilient freight network design. This work contributes to the advancement of data-driven decision-making in transportation logistics, emphasizing the importance of real-time, interpretable, and scalable predictive models in mitigating risk and improving reliability in freight systems.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
