Railway infrastructure requires effective maintenance to ensure safe and comfortable transportation. Among the various degradation modes, track geometry deformation caused by repeated loading is a critical mechanism impacting operational safety. Detecting and maintaining acceptable track geometry relies on track recording vehicles (TRVs) that inspect and record geometric parameters. This study aims to develop a novel track geometry degradation model considering multiple indicators and their correlation, while accounting for both imperfect manual and mechanized tamping. A multi-variate Wiener model is formulated to capture the characteristics of track geometry degradation. To overcome data limitations, a hierarchical Bayesian approach with Markov Chain Monte Carlo (MCMC) simulation is utilized. This study offers a contribution on the analysis of a multi-variate predictive model which considers correlation between the degradation rates of multiple indicators, providing insights for rail operators and new track-monitoring systems. The performance of the models is rigorously validated through a real-world case study on a commuter track in Queensland, Australia, utilizing actual data and independent test datasets. This experimental calibration and validation procedure represents a novel contribution to the existing literature, offering valuable guidance for rail asset management and decision-making.

翻译：铁路基础设施需要有效的维护以确保安全舒适的运输。在各种退化模式中，重复荷载引起的轨道几何变形是影响运营安全的关键机制。检测和维持可接受的轨道几何状态依赖于轨道检测车（TRVs）对几何参数的检查与记录。本研究旨在开发一种考虑多指标及其相关性的新型轨道几何退化模型，同时兼顾人工捣固和机械化捣固的不完美性。通过构建多元维纳模型来捕捉轨道几何退化的特征。为克服数据局限性，采用基于马尔可夫链蒙特卡洛（MCMC）模拟的分层贝叶斯方法。本研究对多变量预测模型进行了分析，该模型考虑了多个指标退化速率之间的相关性，为铁路运营商和新型轨道监测系统提供了见解。通过澳大利亚昆士兰通勤轨道的实际案例，利用实测数据和独立测试数据集对模型性能进行了严格验证。这一实验校准与验证流程是对现有文献的新贡献，为铁路资产管理及决策提供了有价值的指导。