The MiniBond test: A robust and cost-effective alternative for ore hardness characterisation and geometallurgical profiling

Efficient ore hardness characterisation is pivotal for effective geometallurgical modelling and mine planning. However, conventional Bond Work Index (BWi) tests, despite their established accuracy, are resource-intensive in terms of cost, time, and sample requirements. This article introduces the MiniBond test, a novel reduced-scale grindability assay designed to provide reliable BWi estimates using significantly lower sample volumes (approximately 600 grams). Extensive calibration across diverse mineralogical domains (over 80 ore samples) validated a high correlation (R 2 = 0 . 98 ) with the traditional Bond test and internal calibration errors consistently below 3%. Double-blind homologation and validation studies conducted collaboratively between Aminpro and Anglo American confirmed the test’s operational reproducibility across laboratories, achieving less than 5% variation. By significantly reducing the cost per test (approximately 80%) and enabling high-throughput testing, the MiniBond test enhances the density of geometallurgical datasets, improving geostatistical interpolation reliability. This paper presents a detailed description of the MiniBond test procedure, the extensive calibration and homologation campaigns, and discusses the practical implications and advantages of implementing this method for large-scale resource evaluation. • The MiniBond test replicates Bond mill grinding conditions using only 600 g of sample. • Calibration models achieve <3% error versus the standard Bond Work Index. • AATS homologation and blind validation confirmed reproducibility across labs. • Bootstrap learning curves establish minimum calibration dataset size for stable predictions.