Estimating the conditional derailment probability of release (CDPR) of tank cars in freight train derailments

Gharzouzi, Paul

Permalink

https://hdl.handle.net/2142/130122 Copy

Description

Title

Estimating the conditional derailment probability of release (CDPR) of tank cars in freight train derailments

Author(s)

Gharzouzi, Paul

Issue Date

2025-06-03

Director of Research (if dissertation) or Advisor (if thesis)

Gardoni, Paolo

Doctoral Committee Chair(s)

Gardoni, Paolo

Committee Member(s)

• Barkan, Christopher P.L.
• Duarte, Carlos Armando
• Kirkpatrick, Steven W.
• Lin, Chen-Yu

Department of Study

Civil & Environmental Eng

Discipline

Civil Engineering

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Tank car safety
• freight train derailment
• conditional derailment probability of release
• model validation

Abstract

Freight trains are a common mode of transporting goods and products over extensive distances in North America. Tank cars are used for bulk transportation of a wide range of liquid products including both non-regulated and regulated (hazardous) materials. Some hazardous materials (hazmat) that are commonly transported in tank cars include ethanol, crude oil, etc. Despite their low likelihood, train derailments resulting in hazmat releases can have catastrophic human, environmental, and economic consequences. As such, there is a need to adequately assess the safety performance of the hazmat tank cars in a freight train derailment. The conditional probability of release (CPR) is one of the most insightful safety metrics used to assess the safety performance of these derailed tank cars. CPR values of derailed cars were developed from statistical analyses using an extensive historical database. However, the statistical CPR is not applicable to novel tank car designs or materials, or changes in operating conditions due to the lack of relevant or sufficient accident data. Accordingly, a methodology for determining the CPR using analytical or simulation-based approaches is a desirable alternative for these cases. Nonetheless, given the chaotic and complex nature of train derailments, quantifying the CPR of derailed cars through these approaches is quite challenging as it requires a thorough understanding of the dynamic behavior of derailed railcars and information on the derailment impact environment, which is typically unavailable. Information on the impact load environment can be obtained by simulating a series of derailment scenarios that capture the real-world accident conditions. This dissertation develops a probabilistic formulation to determine the Conditional Derailment Probability of Release (CDPR) of novel tank cars under specific derailment conditions. Determining the CDPR is a necessary step to estimate the CPR of novel tank car designs, materials, or changes in operating conditions before they are in service. The probabilistic formulation captures the underlying physics of derailment and impact events through a simulation-based approach. In particular, the formulation proposes probabilistic models to quantify the resistance and demand for tank cars and their key components (i.e., tank heads, tank shells, top fittings, and bottom fittings). These models characterize the structural behavior of each tank car component in an impact scenario. Moreover, these models incorporate key elements, such as the derailment-caused impact forces, impact types, impact event characteristics, and tank car properties, to determine the quantities of interest at the level of an impact type affecting a component of a tank car. As a demonstration of how the formulation works, the CDPR values for three simulated train derailment scenarios, based on the real-world derailment of Graettinger, Iowa, are determined. Focusing on the impact environment in a derailment, this dissertation supports the refinement and applies a high-fidelity three-dimensional (3D) finite-element (FE) derailment model to accurately model a complete train derailment, including the interaction of the derailed railcars with each other and the surrounding environment. The model is able to incorporate a variety of accident and train characteristics, as well as models of different train and site features, such as car and equipment, train braking, and ground resistance. To leverage the output from the FE derailment model, this dissertation develops a comprehensive validation procedure to assess the performance of train derailment models in replicating the kinematics of a full train derailment. By validating the modeled real-world derailment kinematics, the FE derailment analyses can subsequently be used to simulate the derailment impact hazard environment affecting the railcars. To that end, the validation procedure introduces a series of primary and derived metrics to quantify characteristic dynamic behaviors of the train and derailed railcars, such as the extent of the derailment, the spatial dispositions of derailed railcars, their longitudinal and lateral spread, and their misalignment with the track. The procedure also employs standard comparison measures to objectively evaluate the alignment of a modeled derailment with a real-world accident. With that, the kinematic performance of standalone models and that of multiple model iterations, which feature varying model parameters, can be assessed using the developed validation procedure. This procedure is applied to six real-case freight train derailments that occurred in Aliceville, Alabama; Graettinger, Iowa; Doon, Iowa; Cherry Valley, Illinois; Tiskilwa, Illinois; and Crosby, Texas. Overall, this dissertation provides tank car safety experts with a rigorous approach for assessing the reliability of hazmat tank cars, namely novel tank car materials and configurations or changes in operating conditions, under specific accident conditions.

Graduation Semester

2025-08

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/130122

Copyright and License Information

Copyright 2025 Paul Gharzouzi

Owning Collections