Key Takeaways
- AI blasting evaluation splits into two tasks: tracking where the ore actually moved (Hexagon Blast Movement Intelligence) and measuring fragmentation size (Split-Desktop, WipFrag).
- Hexagon BMI generates a “Muckpile Block Model”, a post-blast update to your grade-control block model. It does this without requiring personnel to walk the muckpile to place or retrieve monitors.
- Fragmentation-analysis tools apply image processing and computer vision to muckpile photos. They estimate particle size distribution instead of relying on manual sieving.
- This is a mature AI application. Both categories have years of field deployment behind them, not just pilot studies.
TL;DR
Blasting evaluation with AI answers two questions after every blast: where the ore went and how well it broke. Use Hexagon Blast Movement Intelligence to update your block model for ore movement. Then run image-analysis tools like Split-Desktop or WipFrag on muckpile photos to get a particle size distribution.
What Is Blasting Evaluation?
Blasting evaluation is the review of a blast after it fires. It checks whether the blast did what the design intended. Engineers look at fragmentation (the size of the broken rock), muckpile shape, ore movement, and vibration or airblast. They also note problems such as oversize boulders, flyrock or poor floor conditions.
Two results matter most for the rest of the mine. First, blasting moves ore, so the ore and waste boundaries in the block model shift. Second, fragmentation drives digging rates, crusher throughput and grinding energy. Both used to require manual, labour-intensive methods: movement markers retrieved by hand, and hand-sieved or visually estimated fragmentation.
How Do I Evaluate Blasting With AI?
The work splits into two measurement tasks and one feedback step.
Tracking ore movement after the blast
Hexagon’s Blast Movement Intelligence pairs an AI-driven blast movement engine (developed with Augment Technologies) with Blast Movement Monitors. Walking the muckpile to place and retrieve physical monitors is a genuine safety risk. Instead, BMI’s AI models predict and validate ore movement. The result is what Hexagon calls a “Muckpile Block Model”: your pre-blast grade-control model updated to show where material actually ended up. It integrates with Hexagon’s MinePlan Block Model Manager. As a result, the muckpile model flows straight into ore and waste delineation decisions.
Measuring fragmentation for blasting evaluation
Image-analysis software like Split-Desktop and WipFrag handles fragmentation. Both work from photographs of the blasted muckpile. They run edge-detection and segmentation algorithms to isolate individual fragment boundaries, then compute a particle size distribution. WipFrag also supports zoom-merge analysis. It combines images taken at different scales to overcome the resolution limits of any single photo. For continuous measurement at the shovel or crusher, see our page on AI fragmentation analysis.
Feeding blasting evaluation back into design
The fragmentation curve and the muckpile model are only useful if they change the next blast. Feed them into blast design optimization so burden, spacing and loading adjust to real results. Over time, this loop is what turns a design tool into a trained optimizer.
What to Expect From AI Blasting Evaluation
Expect to pair a dedicated camera or monitor setup with either tool, rather than treating this as pure software. Both categories depend on decent image capture in the field. Results are only as good as your photo quality and calibration reference.
Image analysis also sees only the surface of the muckpile. Therefore, it can miss fines and buried oversize, so most sites calibrate it against sieving or crusher data. Finally, consistent timing matters. Photos taken at the same stage of digging, blast after blast, make results comparable. That consistency also supports better blast sequencing decisions later.
FAQs
What does blasting evaluation measure?
It measures fragmentation, ore movement, muckpile shape, and vibration or airblast after a blast. The two results most used by AI tools are the particle size distribution and the updated position of ore and waste. Together, they show whether the blast met its design goals.
How does AI measure blast fragmentation?
AI tools photograph the muckpile and use edge detection and segmentation to outline each fragment. They then calculate a particle size distribution from those outlines. Split-Desktop and WipFrag are two established examples.
Why track ore movement in blasting evaluation?
A blast can shift ore several metres from its original position. If the block model is not updated, shovels dig ore as waste and waste as ore. Hexagon BMI produces a muckpile block model that corrects for that movement.
Is AI blasting evaluation proven?
Yes. Both blast-movement tracking and image-based fragmentation analysis have years of field deployment behind them. The main limits are practical: image quality, calibration and consistent data capture.
Try It With Geocluster
Choosing between blast-movement and fragmentation-analysis platforms is a multi-tool research question worth doing systematically. So is working out how their outputs should feed back into your blast design. The Geocluster Research Harness helps you evaluate AI tooling choices like this across your blasting workflow. You can also read how the Geocluster Research Harness works.