Data Underlying the MSc Thesis: An Advanced Tool for Evaluating the Probability of Failure of Existing Tailings Dams
doi: 10.4121/c9f8ab0d-ab98-449e-b304-c03c42884826
CC BY-NC 4.0
The frequent occurrence of catastrophic tailings dam failures underscores the urgent need to improve safety practices and minimise associated risks. This study introduces an advanced tool to evaluate the total Probability of Failure (PoF) of existing tailings dams systematically and effectively, supporting the rational prioritisation of mitigation efforts and minimising risk within the ALARP principle.
The tool utilises a semi-quantitative approach, combining observation frequency and expert judgment. A baseline PoF for each dam construction method and failure category is established using a database of 450 tailings dam incidents and accidents developed in this study. This baseline PoF is subsequently modified based on site-specific factors influencing failure prevalence. A total of 255 key contributing factors are identified based on a fault tree analysis, the Global Industry Standard on Tailings Management (GISTM), and experts in the field. The factors encompass site conditions, design elements, and the Level of Practice (LoP), which address all major credible failure modes and mechanisms. The factors are linked to the failure categories they influence and assigned relative weights through the analytical hierarchy processing method for each dam construction method and failure category. Subsequently, the weights are multiplied by modifiers to account for the effects of site-specific conditions (favourable: 0.2, neutral: 1, adverse: 5, and unknown: 2). Within the tool, users can choose fulfilment conditions for each factor from drop-down menus and the selected inputs are connected to the modifiers. The adjusted weights are multiplied by the baseline PoF of each failure category, given the dam construction method. The summation of these products yields the total PoF of the investigated dam.
The results provide preliminary insight into factors significantly affecting the total PoF for the dam under investigation, aiding in evaluating whether the PoF reduction justifies costs. It facilitates preliminary, rational prioritisation of mitigation measures in accordance with the ALARP principle, contributing to ongoing efforts to improve tailings dam safety.
To validate the tool’s capabilities, two case studies with available data are analysed: the Aznalcóllar failure to examine the ability to identify high-risk factors and a recently improved dam to evaluate if the mitigation efforts are adequately reflected. The studies demonstrate the tool’s potential but also reveal uncertainties, inaccuracies, and limitations. These stem from discrepancies in the baseline PoF, weightings, modifiers, and unaccounted factors. Therefore, caution is warranted in the tool’s utilisation. Recommendations include various improvements and further verification and validation across a broader range of case studies. Value can be added by incorporating additional components and adapting the tool for new dams.
Notes:
The tool is developed as part of the Master Thesis: 'An Advanced Tool for Evaluating the Probability of Failure of Existing Tailings Dams', wherein comprehensive details about the development are provided.
I welcome discussions on the tool. Feel free to reach out if you have any points to discuss.
Please be aware of the tool's limitations when using it.
- 2023-12-08 first online, published, posted
RWTH Aachen University (Division of Mineral Resources and Raw Materials Engineering)
Aalto University (School of Engineering/School of Chemical Engineering)
BGC Engineering
DATA
- 1,217,804,078 bytesMD5:
9cbf8d9d931828e911f8c2e3b3b00d5eAn Advanced Tool for Evaluating the Probability of Failure of Existing Tailings Dams.xlsx -
download all files (zip)
1,217,804,078 bytes unzipped
