Sure.
My presentation will be a little different from the previous presentations you had. I was looking at your website. It will be slightly more technical, not because I'm here to say anything else other than to discuss and present, through examples, the effect of technologies on different areas, from digitization, productivity, economic cash flows, economic issues, environmental performance and sustainability. You will be asking, of course, how do these things all connect into one?
Sustainability, and everything that you see on the list, means something slightly different from the general usage. Here sustainability simply means, can I make more out of a mine that I already have or that I am building, based on new technologies?
I thought I should also tell you what the organizational model is that we have been following. I worked for a mining company for some time, spent 10 years in Australia, where I made a lot of friends, particularly in big mining companies. I came back to Canada due to the Canada research chairs program.
In 2006 the idea was that we would set up a consortium. In the companies you see here, from AngloGold, Vale, to BHP Billiton, you probably recognize these very big names. One interesting part here is that most of these, except perhaps Kinross, which joined us last year, are totally non-Canadian. Barrick is also Canadian, of course, but the rest of them are not. I find that interesting because suddenly you see people from Chile, Brazil, Australia, South Africa, coming to fund the work we do.
The budget has been about $1.1 million per year now for five years, so multiply $1.1 million by five. Of course, we capitalize on government programs, particularly the research funding through NSERC, Canada research chairs, of course, and to some extent FRQNT in Quebec.
One thing I find interesting is the way we operate. One thing I found interesting from the beginning, and I appreciate it now more than ever, is the fact that what we set up is not like a professor who has a lab or sets up a lab. It isn't quite that. It is that we have a real collaboration between the companies you see here and the expertise we have in the lab that I run and the new technologies we develop. The whole thing's a partnership. I can tell you extremely clearly that all these companies are in Montreal, at least twice...from anywhere you want.
The key contribution is that they are a think tank. Okay, I'm a professor. Fine. I may know some things as a professor, maybe theoretical things, mathematical, computing, whatever you like it to be, and mining, etc. But what do I know about the large-scale, serious problems these companies have? I will show you an example of the collaboration. When we work we interact consistently, and we become friends until they move on to another position in their company or they retire.
That is, more or less, the model. If you have an interest, you can always explore this more. I say this because I find the way that we operate in Canada to be very different from the way I actually learned in Australia.
I will talk to you about mining complexes. It's not a complex thing; it's just the amalgamation of different things. We start with mines, as you see them here. We continue with the treatment of what we extract. We have, of course, our waste dumps of all sorts and kinds. Of course, the main job we have is to generate products that we sell in the markets.
This is what I mean by “mining complex”. “Mineral value chain” is another term we use. We are having difficulty deciding which one to keep.
Before I continue, the other aspect here is that we realize that managing risk is a serious issue. By “managing risk”, I don't mean anything but “technical risk”. I think the example I have here will show you what I mean by technical risk. I will show you the graph on the lower left corner of the page. It comes from the late nineties, when most of this work started, at the time when I went to Australia and got into these things. What it shows you on the horizontal axis is that the deviation from the expectation of production for these mines is at about 48. Of course, we see deviations from expectations here in one and two that are at minus 60 or very negative, and you understand if that happens that there is no mine.
You also have the other end, whereby some operations may start producing double what was expected. Of course, they are lucky, in that they will not get closed, but at the same time, what very commonly happens in this case is that we set up a structure in a mine that we can not really change that much. We can get stuck with overproduction and a wrongly set-up mine, which generates very suboptimal metal production and cash generation.
An easy way to show you a starting point in all of this is that in mines we have exploration that we call “drill holes”. They're about 30 to 40 metres apart. We model, as you can see in the bottom list shown here, deposits. In other words, we describe them at the scale of mining units, and we interpolate values in these blocks. The values are for metal, but they could be different elements. It could be anything you like that relates to the properties of the materials we extract. I should stress that the materials we extract are extremely heterogenous.
There are new technologies that were started some time ago; I am not the only who works on these. The petroleum industry has used them since the early 1990s. On the one shown here, you can see that we have a mine. You can see three sections. This is a vertical section of a small part of this mine. What I'm trying to show you there is a concept called the “Monte Carlo simulation”, or, as we call it, “stochastics”; it doesn't make a difference to me. The only issue here is that we describe the uncertainty that we see in the deposits that we are exploring. Here, you see, for example, three different scenarios of how the deposit might look based on the information that we have. Different areas from one scenario to the next to the next have different values. Red is high-grade gold and blue is below one gram per tonne.
The key issue here is technologies that can do this. There is a devolution in these technologies from basic to complex. The second thing is that it is not exactly simple to simulate scenarios, as we call them, of mineral deposits when you have a million or two mining blocks describing whatever we'd like to describe.
That opens up, of course, the key question of computational requirements. Of course, if you asked me 15 years ago if I could do two million mining blocks, I would have said that I was not exactly sure how to do it, or I would look for shortcuts. These kinds of things are a starting point to describe technical risk, to say there is now a technical risk so how do I model the deposit and how do I describe gold content if that is the metal of interest?
To go back from that point—