Awards Profile: Orica’s design for the result

Orica’s Design for Outcome (DfO) applies machine learning algorithms to automatically characterize rock for blasting, setting a new benchmark for blast design in the digital age. It unleashes the power of using previously underutilized data sources upstream and downstream of the blast to design the best blast every time.

By automatically processing data to better understand rock mass variability and adapting blasting designs accordingly, more consistent and targeted blasting results are achieved. This improves productivity and reduces costs throughout the mining value chain.

SANDING ENABLED BY MACHINE LEARNING – THE OPPORTUNITY

Today, blasting generally produces variable results. This negatively impacts downstream productivity and overall costs due to lower fill factors and shovel productivity, oversizing, reduced crusher throughput and crushing productivity scaled down.

The energy of the explosives does not correspond to the variable properties of the rock. Despite an increase in sensor data available in mines, conventional blast design techniques make it virtually impossible for an engineer to combine and process all of the data sources needed to adapt blast designs to high-resolution rock conditions. . More importantly, to make it into a daily workflow of mining operations. Design for Outcome applies machine learning and automation to the problem, creating a solution that allows drill and blast engineers to practically implement custom data-driven blast designs.

FUNCTIONALITY

Design for Outcome uses machine learning algorithms and cloud-hosted processing to match rock characteristics to high-resolution blast energy for targeted and consistent results. It also automates the process and presents the user with a simple web interface that makes it easy to generate high-resolution bespoke sandblasting designs. Design for Outcome comes in a variety of configurations: • Post-Drill Classification Module This module automatically ingests Measure While Drill (MWD) data and produces hardness-based dynamic domains.

The generated output is used in blast design software packages with loading rules capability, such as Orica’s SHOTPlus™ Premier blast design application.

• Pre-drill classification module

The pre-drill classification module uses block model data supplemented with existing high resolution MWD data. Machine learning algorithms generate explosibility domains that inform drilling patterns for subsequent explosions. The field of dynamization is then refined after drilling. The raw MWD is cleaned and normalized to produce domains for automated loading rules using the post-crawl classification module.

BENEFITS

Designed in collaboration with Orica customers, subject matter experts and data scientists, the technology can be used to efficiently allocate drilling and explosives energy for consistent results, providing the following benefits:

• Reduced drilling and blasting costs while maintaining productivity by eliminating excessive use of explosive energy

• Improve downstream fragmentation and productivity by allocating proper energy to harder rock domains

• Use of data from sensors implemented in mines in operational workflows rather than retrospectively

• Efficient generation of custom blast designs within production schedules despite the use of high volumes of data

• Ease of use via a web interface for drill and blast engineers CASE STUDY ROY HILL Through an innovation partnership, Design for Outcome was deployed at the Roy Hill iron ore mine in Western Australia to improve mine production and increase mining profits by reducing variability in blast performance.

Working with Roy Hill, Orica integrated blast drilling design data into a cloud-hosted platform, providing data analytics to generate valuable insights into geological hardness, energy deployment and productivity mining.

Design for Outcome machine learning algorithms use MWD data from the autonomous drill fleet to define the borehole geology domain and match the energy of explosives to each domain, generating automated loading rules for the Blast loading and deployment in the field through intelligent delivery of explosives enabled by BlastIQ™ units.

The closed-loop system comes with an automated excavation productivity report. By using Design for Outcome, Roy Hill was able to target higher energy only where it was needed, dramatically reducing drill and blast designs and costs throughout the mine while maintaining mine productivity. reference excavation.

AND AFTER

Design for Outcome will continue to increase value creation by better targeting drilling and explosives energy for cost and productivity benefits.

This includes:

• use new and emerging sources of ore body knowledge data to better illuminate the high-resolution domain

• integration of explosion results data sources into machine learning algorithms

• Advance end-to-end integrated capability, from ore body knowledge to process optimization, through integration with technologies such as Integrated Mining Simulator (IES) to enable sustainable Mine to mill.

Abdul J. Gaspar