OFFICIAL • As-Drilled Stage: Once drilling is completed (and actual drill logs are available), the model is rerun using updated inputs (actual hole depths, etc.). This second-stage validation refines the clearance requirements and can result in an adjustment of the charging requirements. • As-Charged Stage: Once explosive loading is completed, the model is again rerun using updated inputs (actual explosive mass, stemming, etc.). This third-stage verification refines the hazard estimate and can trigger last-minute changes, such as relocating equipment or personnel to a safer distance if the risk appears higher than initially thought. To standardize across all the company’s mine sites and accommodate the adapted prediction model, a web application (FRED) was developed (Figure 7). The application is linked to the drill and blast design and management software used by Anglo American’s mine sites, namely BlastLogic (Tordoir and Roberts, 2019). Figure 7 – Web application developed for flyrock prediction and projectile database 3.2 Monitoring Component The capability to monitor and quantify actual flyrock events is a fundamental element of the framework. Van der Walt and Spiteri (2020) highlighted key limitations in existing measurement approaches used to validate prediction models, concluding that more accurate, reliable, and unbiased methods are required to precisely determine flyrock travel distances. McKenzie et al. (2015) further noted that continuous improvement of blasting practices represents a key preventative measure against flyrock, relying heavily on systematic video capture, secure storage, and structured analysis and classification of blast footage. The routine collection of data from flyrock events, even where no incident occurs, has been shown to significantly enhance the accuracy and robustness of existing predictive models (Zhou et al., 2024).
RkJQdWJsaXNoZXIy MTM0Mzk2