Geological uncertainty presents a significant challenge in mine-planning. While infill drilling can reduce this uncertainty, identifying the most valuable drilling locations remains complex. This paper presents a chance-constrained optimisation approach that integrates drilling decisions into an ore blending model. By incorporating uncertainty directly into the planning process, the model guides drilling toward areas where reducing variance most improves blending outcomes. Chance-constrained programming enables the blending of uncertain geological properties to maximise plant output while maintaining quality constraints within acceptable risk levels. The model quantifies the value of additional drilling, prioritising locations that enhance blending flexibility. A case study from a coalmine in Queensland, Australia, demonstrates that targeting high-quality blocks for variance reduction yields better results than traditional approaches focused on low-quality areas. This method offers a practical and effective strategy for optimising mine plans under uncertainty and highlights the value of probabilistic modelling in resource planning.