Gamboa, U & Janzon, A 2025, 'Use of intensive remote monitoring and artificial intelligence
to optimise paste pump availability on paste fill', in AB Fourie, A Copeland, V Daigle & C MacRobert (eds), Paste 2025: Proceedings of the 27th International Conference on Paste, Thickened and Filtered Tailings, Australian Centre for Geomechanics, Perth, pp. 303-314, https://doi.org/10.36487/ACG_repo/2555_21
(https://papers.acg.uwa.edu.au/p/2555_21_Gamboa/)
Abstract: It is widely recognised that a conventional paste fill plant layout incorporates several pieces of equipment used to build and dispose of cemented paste in the underground mine. From all the related main equipment considered in a conventional paste plant, the pumping system plays a key and crucial role for paste pouring underground on a safe and controlled way through a high-quality paste that needs to meet the unconfined compressive strength (UCS) required underground based on the design to meet the mining cycle.
This paper as a continuation of a previous paper presented by Gamboa & Castilla in 2024, aims to analyse and study how continuous monitoring of a piston-diaphragm type pump and the usage of artificial intelligence (AI) applied to the vast amount of data collected, has enabled the entire paste plant to achieve availabilities above 97% by substituting conventional preventive or corrective maintenance practices with a predictive procedure. Exploiting the data collected by the pumping system applied to paste fill will provide a good opportunity to create intelligent systems in the near future, as some works have been done already with a similar principle to predict UCS underground (Mariem et al. 2024).
This system offers clear benefits on the pump and therefore paste plant availability and reliability, while enabling a smooth maintenance and paste production planning, with an optimum binder dosage and mining cycle synchronisation.

Keywords: paste pump, paste fill monitoring, artificial intelligence