Reliability-Based Optimization of Rail Inspection

This thesis proposes a quantitative method to optimize inspection/repair intervention in rail defect management. Rail defect management is important for track maintenance, since rails are the most significant and basic component of the track systems. Rail inspection is a fundamental intervention to prevent rail failure. Railroads have evolved the rail inspection interval based on their empirical judgement and on field data. A crack size is predicted by linear elastic fracture mechanics (LEFM). The proposed method identifies the time-varying rail reliability due to deterioration, using data obtained from LEFM and from first-order reliability methods (FORM), which consider the uncertainty regarding the model. Since FORM is an approximation method, Monte Carlo simulation confirms the results. To represent practical situations regarding rail defect management, an event tree (ET) analysis is performed. The ET is modeled to all events and actions with respect to inspection/repair intervention. The ET analysis evaluates the expected reliability of a rail after inspections and possible remedial actions. Based on these results, a life-cycle cost (LCC) model is formulated, taking into consideration the time value of money. To this end, applications of the model to optimization of inspection intervals and to investigation of the effect of nondestructive testing and remedial actions on the LCC and the interval are analyzed. As a result, it is possible to extend the present inspection interval. Additionally, the effect of detectability of nondestructive testing on the LCC is more significant than that of accuracy of nondestructive testing, and a proactive maintenance policy may reduce both the expected total cost and the number of inspections. Thesis Supervisor: Jerome J. Connor Title: Professor of Civil and Environmental Engineering