Open Collections

Pilbara Case Study : Tailings Dams’ Failure Probability Quantification, Worlds’ Benchmarks and ALARP

Tailings and Mine Waste Conference (2023 : Vancouver, B.C.); University of British Columbia. Norman B. Keevil Institute of Mining Engineering

This paper presents the quantitative probabilistic failure analysis of a Pilbara active tailings dam. The case was proposed by Ryan Singh as a round robin exercise, focused on a limited failure mode selection. Due to confidentiality, we generally cannot disclose the sites we use as examples, which makes this case an exceptional opportunity, as a cross section and various drawings are publicly known. Despite numerous initial objections related to the design of the exercise, the authors decided to participate, as we have performed to date more than hundred quantitative risk assessments of active/inactive dams all over the world using the same methodology, and these have proven to mimic the behaviour of the world-wide portfolio. Thus, it was interesting to see the influence of the round robin limited scope of failure modes on the overall probability of failure, and then to show how this dam would compare to the worldwide portfolio. This paper first discusses a quantitative approach that delivers an order of magnitude estimate, based on the dams’ similarities with respect to the world-wide portfolio. This is an approach that does not require any numerical analysis and is based on common sense; however, it is very approximate. The paper then delves into a more sophisticated quantitative probabilistic reliability approach. The full array of results is shown and explained for the specific dam, including the round robin selected failure modes and a more complete list. These results are then inserted into the world-wide portfolio and benchmarks system we have built, to show how this dam compares with the others we have studied, including the influence of arbitrarily selecting failure modes versus a complete analysis. Finally, the paper closes with a discussion of possible climate change impacts and shows how the ALARP analytical quantitative definition can be sought.

Text

2023-12-08

Attribution-NoDerivatives 4.0 International

http://hdl.handle.net/2429/86786

Unreviewed

http://creativecommons.org/licenses/by-nd/4.0/