Abstract
The mining industry faces significant challenges in managing both inactive and active tailings storage facilities (TSFs), which vary widely in size, environmental parameters, and maintenance issues. Traditional risk assessment methods for TSFs often rely on incomplete and inconsistent historical data, leading to potential misclassifications and a false sense of security. This paper introduces a comprehensive quantitative risk assessment model for TSFs that addresses these limitations by incorporating human factors and evolving risks due to maintenance, operations, and climate change. The model leverages a probabilistic causality analysis to evaluate the failure processes of tailings dams, emphasizing the importance of human factors over natural events in determining failure probabilities. By analyzing a global portfolio of dams, the model provides a robust framework for assessing and mitigating risks, offering practical insights into the effectiveness of various mitigation strategies. Comparative analyses of three similar dams—categorized as “The Good, Bad, and Ugly”—demonstrate the model's capability to differentiate risk levels based on care and understanding in design, construction, and maintenance. The paper also explores the potential for preparing scripts to study mitigative alternatives, such as buttress construction or less significant ones, and discusses the application of ALARP (As Low As Reasonably Practicable) analyses in optimizing risk management. This work aims to enhance the reliability and safety of TSFs by providing a more accurate and holistic approach to risk assessment.
*This paper was named among the top 5 papers at the Dam World 2025 Conference. Read the press release here for more information.
