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
Random Forest-Based Covariate Shift in Addressing Non-Stationarity of Railway Track Data
-
2023-09-06
[PDF-685.23 KB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:United States. Department of Transportation. University Transportation Centers (UTC) Program ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology ; University Transportation Center on Improving Rail Transportation Infrastructure Sustainability and Durability. USDOT Tier 1
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:For years, track geometry vehicles have been deployed to capture rail defects. However, the limitation associated with the operation is the possibility of non-stationarity of the observed measurements due to external influence. The effect of non-stationarity may lead to the false representation of track conditions and thereby increases the likelihood of false output. For that reason, we considered the possibilities of supervised machine learning techniques for detecting and correcting the track geometry inherent anomalies. The methods include Random Forest (R.F.), Logistic Regression (L.R.), and Support Vector Machine (SVM). To ascertain the discrepancies within the data, we varied the train-test and validation ratio in phases. Conclusively, the developed models’ application indicates that the Random Forest is a more practical approach to detecting the non-stationarity of track geometry data. Also, it optimizes the cost of maintenance and supports accurate decision making to improve track safety better.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
