OFFICIAL DEVELOPMENT AND IMPLEMENTATION OF A QUANTITATIVE FLYROCK FRAMEWORK FOR SAFER AND MORE PRODUCTIVE SURFACE MINING *E. Carpenter1 1 Anglo American Plc, Johannesburg, South Africa (*Presenting author: ernest.carpenter@angloamerican.com) ABSTRACT Critical minerals transition has increased reliance on large‑scale surface mining. As a result, drill and blast operations face growing pressure to consistently produce diggable material while meeting stringent safety, environmental, and productivity requirements. Among blasting hazards, flyrock continues to present a high consequence risk, yet its management is often reactive by nature. Flyrock can cause fatalities, damage critical infrastructure, and trigger costly production stoppages. When proactive management does occur, it is generally fragmented across prediction tools, left to ad hoc observations, and reactive to incident investigation. This paper presents the development and implementation of a Quantitative Flyrock Risk Management Framework, designed to integrate modelling, monitoring, measuring, and management into a structured process over the full drill and blast cycle. The modelling component explicitly adopts site and domain specific parameters. Blast design information, geological domains, execution compliance, and measured fragment morphometry data are systematically captured and used to calibrate trajectory inputs, improving the credibility of predicted flyrock ranges. Modelling is applied at multiple stages ensuring that local conditions and execution variability are directly reflected in predicted exclusion distances. Monitoring is structured around systematic blast video capturing which is supported by recent advances in video analytics and blast monitoring technologies to improve the efficiency, consistency, and objectivity of post blast reviews. The measuring component converts observed blast outcomes into quantitative data by extracting projectile trajectories and impact locations from video footage and digital terrain models, enabling direct comparison between predicted and actual flyrock behaviour. These results feed into the management component, which incorporates structured reconciliation routines where predicted risks, observed outcomes, exceedances, and contributing factors are formally reviewed and corrective actions developed. Through this process, a structured flyrock database is progressively established, providing a reliable foundation for application of data analytics and artificial intelligence to further refine prediction accuracy and decision making. Roll out of the framework at five Anglo American operations has ultimately resulted in an uplift in flyrock risk management practices and awareness. Early-stage adoption confirmed flyrock exceedances, beyond the equipment radius, occurred far more often than the formal incident count suggested, providing the motivation for better practices. The practicality and ease of embedding the framework into operational workflows are demonstrated through the digitisation of critical operational data and the leveraging of enabling
RkJQdWJsaXNoZXIy MTM0Mzk2