Rail corrugation is a widespread undulatory rail wear on the rail running surface. Untreated rail corrugation can lead to many adverse consequences. To reduce the growth of rail corrugation, several experimental studies were conducted to determine the causes and contributing factors of rail corrugation, with little emphasis on the analysis and prediction of corrugation growth, which is vital for improved maintenance productivity and enhanced asset management. This paper, therefore, aims to model and predict rail corrugation growth, as well as simulate and analyze axle box acceleration (ABA) signals, for four different track configurations with dissimilar rail metallurgies, sleeper distances, and track systems. By means of three-dimensional finite element modeling and signal processing, results consistently demonstrate that rail corrugation grows exponentially over time, with the slowest growth rate for a ballasted track with 700 mm sleeper distance and head-hardening rails. Under such track configuration, rail maintenance activities are suggested before every 1359565 wheelset passages to achieve compliance with international standard ISO 3095:2013. Furthermore, an increase in wear depth can elevate wheel vibration, raise ABA signal amplitude and power, decrease dominant vibration frequency, and increase dominant corrugation wavelength, which facilitate automated continuous monitoring of rail corrugation growth and development. The present results substantially contribute to the growing body of recommendations for the design, construction, and maintenance of reliable and safe railway systems.
