The drilling and blasting industry is on the verge of unprecedented transformation. New developments in technology have provided all the building blocks for more automated and self-optimizing operations. Now, the biggest challenge is putting them all together.
The opportunity presented for assembling these blocks is complex as it spans technical, commercial and cultural considerations. The ability to plan and deliver on these three considerations will separate the winners from the losers in the new era of digitized mining.
In this Webinar we break down our learnings from 100+ customer interviews and observing the digitalization trends from 600+ sites.
- what the future of Drilling and Blasting could look like
- The technologies that will enable this future
- The current state of tech adoption as well as near term trends
- How current tends could or already are impacting your business
- Major blockers and examples of how some companies are overcoming them
Strayos COO and blasting expert Brad Gyngell to explore the state of digitalization of the Drilling & Blasting industry and how to take advantage of these new developments to increase your operation's sustainability and stay ahead of the competition.
Q & A
We'll start at the end...
Unfortunately we ran out of time before we were able to answer everyone's questions.
So, as promised, here are the answers to all the questions asked during the webinar:
Can you point us to a site from which the McKinsey Paper can be obtained?
Here's the BCG report - Racing Towards a Digital Future in Metals and Mining https://www.bcg.com/en-au/publications/2021/adopting-a-digital-strategy-in-the-metals-and-mining-industry
Here's another report from this year from ABI Research estimating a 9.3 Billion investment in digitalization for mining: https://finance.yahoo.com/news/mining-industry-forecast-embrace-digitalization-080000231.html?guccounter=1
In your studies for your models or designs do you have the use of NONEL versus Electronic firing of shots?
Yes models can be used to quantify the benefits and costs of using non-electric vs electronic initiation. Generally, studies show that the reduction in delay scatter improves fragmentation enabling pattern expansion up to ~10%. In large mines this almost always justifies the switching cost to electronics. The impact however will be unique to each rock type and mining method so it is important to develop a model using site specific data to best evaluate this decision.
Will the push into D&B 4.0 involve a heavy emphasis on data verification, (quantity, validity, lack of bias, etc.) given the crucial role it plays in the AI decision making? Avoiding the Garbage In, Garbage Out, trap.
Yes absolutely, the development of an integrated set of data sources will bring more importance to data validation. In the D&B 1.0 and 2.0 world there is still a chance for manual human verification and cleaning. To unlock continuous learning models and autonomous operations however, this data validation will also need to be automated.
This will require a centralized approach to data governance where standards for data formats and conventions must be defined at the platform layer. Basically, we're going to have a lot of things talking to one another so they're all going to need a common language.
Can AI involvement make it possible to speed up what could be called "consequence analysis"? For example, if we don't actually fire today, (bad thing) will that make it possible to fire a great blast tomorrow? (a very good thing). I ask because, sometimes poor decisions can be made due to short term focus, too short term.
The D&B 4.0 world will enable this type of real-time decision making. This will be unlocked by models predicting the implications of our operational decisions, as well as real-time data flowing into them.
Any comments or thoughts on the role of the tremendous importance of "KPI Setting 4.0" to ensure full cultural acceptance?
"KPI Setting 4.0" is exciting because it represents a self-reinforcing feedback loop bringing major value to mining operations. Better operational data analytics will enable more value based KPIs; value based KPIs will then promote a culture of digital adoption and thus better analytics.
For example, getting quantified data on digging performance for each shot enables D&B KPIs around loader productivity. Once these are in place, D&B SIs, crews and engineers will be more incentivized to implement design optimizers which will result in the gathering of even more data.
This KPI setting will be one major factor driving cultural acceptance, however full cultural engagement will also require strong leadership as well as structural changes for many organisations.
'MVP' means Minimum Viable Product. This refers to the lightest possible version of a technology/application/initiative that still creates some value. Generally, this refers to a solution with a limited scope or feature set that is used to prove the value of the full solution.
Starting with an MVP improves the speed to output and ensures a focus on value. It helps prevent situations where a multi-year development results in the launch of a product that has a lot of features but doesn't properly address the main problem it is there to solve.
'AI', means different things to different people. Is this AI model just an statistical tool doing multivariate analysis? If not, how does the AI work?
AI involves the programming of algorithms that "mimic" human thinking such as learning and problem solving. The algorithms are Programmed to operate within a data set,
identify patterns in that data set and then from those patterns make predictions about what would happen if new data were entered. The algorithms can be Trained to achieve a desired result by testing different combinations of variables to determine which combination will accomplish the desired result and test every possible combination at a speed of millions per second identifying the combinations that work.
There will be multiple types of AI models required to enable a D&B 4.0 world some of which will include multivariate analysis and some of which will not.
The mine to mill optimization model will involve some multivariate analysis but will leverage machine learning techniques which can enable it to handle many more variables at much faster processing speeds.
The fragmentation AI models however do not function in this way as they are more similar to image classification models.
Many sites work off rule based blast designs. In order to collect enough data for the AI to start building an algorithm, do you have breake these rules to design blasts with a wider range of blast design parameters?
Like all statistical techniques, AI will be most accurate within the range of data that has been tested. It is possible however, to implement it before you have built these extended datasets using a range of different techniques.
For example, hybrid empirical/AI models allow you to get a starting point using generalized models which can then be calibrated to your site conditions using AI. The advantage of this approach is that the models can still function outside of the tested range of your sample data due to the empirical relationships established through research.
65% of mines globally have at least one project about IoT and AI, but I am not aware of fully connected mines. What is your opinion about how much time takes to this transformation?
This digital transformation to a fully connected mine could take 2-3 years with full commitment from leadership. It is likely however that mines will execute more slowly than this until precedent has been set and there is a competitive necessity to achieve it. This estimation is based on other industries such as energy, banking and telecommunications who are going through similar transformations but slightly further progressed.
How do you measure efficiently blast fragmentation (including oversized) ? Thanks.
Blast fragmentation can be measured in two main ways
- physical sieve analysis (more accurate but less efficient)
- image analysis (less accurate but more efficient)
Physical sieve analysis is too impractical to conduct on a regular basis and so image analysis is the predominant method of fragmentation analysis. It is also the only method which will be able to feed the real-time data requirements of a D&B 4.0 world.
Within image analysis there are multiple options
- drone camera
- handheld camera
- calibrated belt camera
- equipment camera
The most efficient of these methods is using drone images as it only requires a few minutes after the blast and does not delay loading operations. The drone being used to film the blast can fly straight over and take ~50 photos. These can then be uploaded for processing and the fragmentation measurement will be ready in ~30 min. In Strayos, this will then automatically be used to calibrate fragmentation prediction models for future blasts.
To provide more detail on the body of the shot, the dig face can be re-flown at regular intervals or the drone data can be supplemented with data from other sources.
Poll Questions & Voting
Brad helps mining companies maximize the value of their operations. He spent the first phase of his career helping over 100 sites optimize drill and blast activities as a mining engineer with the world's leading explosives providers Orica and Dyno Nobel.
He then expanded the scope of his support to encompass all commercial and operational functions by moving into management consulting with the top tier firm Boston Consulting Group. As COO of Strayos, Brad leads the development of AI solutions that combine visual and operational data sources to solve tangible use cases for mine operators.
New technologies are rapidly changing the drilling, blasting, mining, and aggregates industries, empowering them in ways never before possible. Make sure you are taking advantage of the best tools available.
Check out our 2 Free E-books on AI applications for the drilling, blasting, and mining industries to see all the amazing advances that are available.
AI Guide for Drilling and Blasting
AI Guide for Mining
Want to watch our videos instead?
Get the latest posts delivered right to your inbox