Railway Track Quality Assessment Using Topological Data Analysis

The US railway system requires continuous monitoring to ensure safety and efficiency. Traditional track quality indices (TQI) often fail to detect variations in track geometry, missing critical irregularities that impact safety and ride quality. This paper explores the use of topological data analysis (TDA) to enhance track quality assessments, introducing a topological track index (TTI), which incorporates persistent homology to capture topological features in track geometry data. Using US railroad inspection data, we compare TQI and TTI metrics across track segments. Results show that TTI is more sensitive to both large and small fluctuations in track geometry, identifying irregularities that TQI overlooks. Results show substantial quantitative differences between the two indices, with TTI values differing from TQI by up to 200% and more. The paper demonstrates TDA’s potential to improve railway maintenance practices, offering more accurate tools for predictive maintenance, track repair prioritization, and enhanced infrastructure integrity.