Baghbani, N & Baumgartl, T 2024, 'Predictive modelling of slope reliability for a Victorian open pit mine using numerical and artificial intelligence techniques', in AB Fourie, M Tibbett & G Boggs (eds), Mine Closure 2024: Proceedings of the 17th International Conference on Mine Closure, Australian Centre for Geomechanics, Perth, pp. 85-98, https://doi.org/10.36487/ACG_repo/2415_04
(https://papers.acg.uwa.edu.au/p/2415_04_Baghbani/)
Abstract: In mining operations, ensuring the stability of slopes is paramount for safety and operational efficiency. This study focuses on predicting slope reliability in open pit mines, utilising a combination of numerical modelling techniques and artificial intelligence (AI). The case study centres on a lignite mine in the Latrobe Valley, Victoria, incorporating key geotechnical parameters such as overburden thickness, lignite strength properties, and slope angle to develop predictive models. Using approximately 30 datasets, both linear and non-linear AI models were developed to generate predictive equations for slope reliability. The linear model achieved a coefficient of determination (R2) of 0.832 for the training dataset and 0.762 for the test dataset, while the non-linear model demonstrated even higher precision with R2 values of 0.963 and 0.929, respectively. This study underscores the critical influence of slope angle and cohesion on slope reliability, offering valuable insights for the management of open pit mining operations. By refining predictive modelling techniques, this research contributes to enhanced safety protocols and operational effectiveness in the mining industry.

Keywords: Factor of Safety, artificial intelligence, open pit mine, Victoria