A unique, ground-breaking software to predict and control dilution and losses in underground mines.


Want to receive more information or speak with a specialist? Sign-up using the form below and a specialist will get in touch with you!

What are the possible returns ($)?

  • Say your mine produces 1,000 tonnes per day (365,000 tonnes/year).

  • You estimate that your average dilution is about 15% (54,750 tonnes/year).

  • Your cost of exploration is 100$ per tonne.

  • Every 1% of reduction of dilution represents savings of 365,000$/year.

  • We estimate that SmartDIL can help you reduce dilution by 2 to 3%, which represents savings between 750,000 to 1,000,000 $ per year.

What is dilution and why is it important?

  • During stope mining, there is always some barren rock that gets blasted with the ore.

  • This waste rock will dilute the ore grade and will add unecessary costs and more processing time.

  • More dilution means more tailings, which will have to be added on the surface; thus, increasing the ecological footprint of the mining operation.

  • It is difficult to understand the exact cause of dilution and to predict with precision.


  • Increased profitability
  • Decreased Time to find an optimal solution.
  • Simplified web interface and detailed reports
  • Reduced ecological footprint


SmartDIL is an artificial intelligence tool for predicting and controlling dilution and losses in underground mines. Our mission is to make underground mines more profitable by providing an easy-to-use artificial intelligence dilution tool and to reduce the mining ecological footprint by understanding the industry’s challenges.

What is Artificial Intelligence?

Artificial Intelligence (AI) is a collection of system techniques that can detect or predict results based on given data. These results can be obtained by humans through an iteration process; AI mimics human-thinking and performs automation to derive those results, allowing organizations to save an immense amount of time.

With many existing models, “Neural Network” is one of the most common AI models out there.

What is the role of humans in AI?

Feed the AI with data

Predict results using AI

Output results from AI

Make decisions based on results

How it works

Data Analysis using AI

SmartDIL collects and normalizes historical site data. Examples include dimensions, rock type, drilling-and-blasting, etc.

Dilution Prediction

After planning your stope, you will simply input the parameters and SmartDIL will predict the dilution of the planned stope based on the extensive collected data.

Design Modification

Using an easy-to-use web interface, you can modify the parameters and SmartDIL will give you an instant analysis of the planned stope’s dilution.

Stope Mining

You can now adjust the planned stope according to the new parameters and perform drill-and-blast.

Add to Secure Database

All the new results from the blasted stope will then be input in a secure database, which will be used for the next stope planning.
Interested in learning more about dilution from experts? Check out the latest episode on The Mining Experience where Mr. Yvan Dionne and Mr. Kilian Bao discuss the importance of underground mining dilution and how to effectivley manage it to reduce on costs.
Interseted in learning how to use SmartDIL? The SmartDIL Webinar Recording is now available! Click the video below to watch the full presentation.

How long does the process take?

Achieving a level of complete automation with minimal manual intervention can take up to 6 months, due to the learning curve of the AI based on stope data.

Once this level is achieved, AI will be completely integrated into the underground mine design and no longer considered as a separate technology.

This system is not static, giving you the value of improved processes that will allow you to improve dilution control continuously based on new insights.

What are the advantages of SmartDIL?

  • Web application – can be setup online and accessed from anywhere

  • No expensive equipment needed.

  • No need much experience in AI knowledge, the program takes care of that!

  • No need to assign others who work in other domains and are outside of this project’s domain. One key employee is needed to input the data and interpret the output.

  • More productivity. Eliminate time-consuming and repetitive tasks done by humans. New insights generated by the AI allows for better decision-making.

  • Environmental sustainability. Less dilution = Less hauled tonnage = Less equipment fuel and wear.

  • Continuous improvement. By continuously collecting stope data, the AI gets better over time.

Predictions accuracy

The main parameters influencing the accuracy of the prediction:

  • The number of mined sites given to AI. This can be related to the number of stopes as well as the accuracy of the results of each mined stope.

  • The other parameters having an influence (not necessarily in this order):

    • Quality of the extracted data.

    • Site with similar properties.

    • Adjustment of the weight of certain parameters by the users.

    • Varieties of extracted data: blasting data, geological data, field support, etc.

  • The accuracy of the prediction will be greater with a greater number of mined sites.

  • The accuracy of the prediction can be represented with the following curves :