Modeling rail fatigue behavior with multiple hazards

Derailments can be caused by rail fatigue failures, which are of great concern to railroad companies. These companies, therefore, have as an objective to predict the locations and times of potential failures and to perform preventive maintenance. Understanding the relationship between fatigue failures and factors affecting fatigue is essential to estimate when and where rail fatigue failures will occur. This paper presents a statistical rail fatigue model in four stages: 1) the model describes the relationship between probability of failure and usage; 2) the model extends to cases of multiple types of failures; 3) the model is derived for discrete usage categories or time intervals with a continuous spatial representation of rail units; and 4) explanatory variables capturing geometry, maintenance, and material properties are included to explain differences in fatigue behavior among heterogeneous rail units. Using available rail fatigue data, the model is estimated. The modeling framework is useful for other multitype failure behavior and for reliability problems, such as vehicle breakdown.