By the way, everyone should know that a few years ago, I made a comment about the font size. My chairman's agenda is about 30 pages long now. Thank you. Good afternoon, ladies and gentlemen, and welcome to Dundee Precious Metals' 36th annual meeting. I now call the meeting to order. My name is Jonathan Goodman, and as Chair of the company, I will chair today's meeting. I'd like to introduce the other directors present today, and I would ask that they stand as I do so. Rick Howes, President and CEO of the company, Peter Gillin, Jeremy Kinsman, Juanita Montalvo, Peter Nixon, Marie-Anne Tawil, Tony Walsh, and Don Young. During this formal portion of the meeting, only registered shareholders present and in person and holders of valid proxies are permitted to participate in the voting.
Following the formal business, Rick Howes will introduce his executive team, provide an update on the business strategy and finances of the company, and respond to questions. Shareholders and guests are welcome to participate. Appointment of Secretary and Scrutineer. Kelly Stark-Anderson, Vice President, Legal and Corporate Secretary of the company, will act as Secretary of the meeting. We've got to update that, Kelly. She's actually Executive Vice President now. The list of shareholders entitled to vote at this meeting has been tabled by Computershare Investor Services as representatives Daniela Munoz and Bryce Dougherty are acting as Scrutineers of the meeting. Has everyone entitled to vote registered with the Scrutineer? If not, please do so now. They're at the back of the room. This notice calling this meeting was mailed to shareholders of record as of March 22nd, 2019.
Declaration for the mailing is available for inspection by any shareholder and will be retained with the records of the company. There is a quorum present, and as notice of the meeting has been properly given, I declare that this meeting has been properly called and is regularly constituted for the transaction of business. The Scrutineers have advised that the proxies received have been voted in support of all of the matters on today's agenda. Accordingly, voting will be by a show of hands. The minutes of the previous shareholders meeting of the company are available for review by any shareholder. Unless someone wishes to have them read, the minutes are tabled and will be retained with the records of the company. Are there any objections to that procedure? You've all received a copy of the circular, which outlines the business for today's meeting.
The consolidated financial statements of the company for the year ended December 31st, 2018, together with the auditor's report to shareholders are tabled and copied, have been mailed to requesting shareholders. There are additional copies available at the entrance to this meeting room. PricewaterhouseCoopers LLP, the auditor of the corporation, is represented here today by James Lusby and Omar Salas. Are there any questions on the financial statements of the company for the year ended December 31st, 2018? The next item of business is the election of directors. As outlined in the circular, management nominates Peter Gillin, Rick Howes, Jeremy Kinsman, Juanita Montalvo, Peter Nixon, Marie-Anne Tawil, Anthony Walsh, Donald Young, and myself, Jonathan Goodman, as directors for the ensuing year or until our successors are elected or appointed.
In accordance with the advanced notice bylaws of the company for nominations of directors by shareholders, we did not receive any notice of any director nominations for this year's meeting. Accordingly, the only persons eligible to be nominated for election as directors of the company are the persons nominated by management. May I have a motion nominating these persons as directors of the company for the ensuing year? Thank you, Shantel. Is there a seconder? Thank you. I will now put the motion to the meeting and call for a vote by a show of hands for the management nominees. For the motion? Against the motion? Carried. I notice that she gets to say the people whose names have been read, but I got to read them all out every time.
I declare Peter Gillin, Jonathan Goodman, Richard Howes, Jeremy Kinsman, Juanita Montalvo, Peter Nixon, Marie-Anne Tawil, Anthony Walsh, and Donald Young, elected as directors for the ensuing year or until their successors are elected or appointed. The next item of business is the appointment of the auditor of the company for the ensuing year. May I have a motion for the appointment of the auditor for the ensuing year? Do I have a seconder? Thank you. I now put the motion to the meeting and call for a vote by a show of hands. For the motion? Against? Carried. I declare the motion carried and that PricewaterhouseCoopers LLP, the chartered professional accountants, is appointed auditor of the company for the ensuing year. Accepting the company's approach to executive compensation. The form of this resolution is outlined in the circular and must be passed by a majority of votes cast at the meeting.
I will now ask for a motion to pass the non-binding advisory resolution accepting the company's approach to executive compensation. Thank you. Is there a seconder? Thank you. I now put the motion to the meeting and call for a vote by a show of hands. For the motion? Against the motion? Carried. I declare the motion carried, and the non-binding advisory standard resolution is approved. This concludes the formal business of the meeting. There is no further business to be brought before the meeting, may I have a motion to terminate the meeting? Is there a seconder? Thank you. I now put the motion to the meeting. For termination? Against? I declare the motion carried. This concludes the formal business of the meeting, and I declare the meeting terminated. I will now turn the meeting over to your CEO, Rick Howes.
I'm going to go and sit in the audience because I want to watch his presentation.
That was me. We have decided to two new members. Part of the executives that like the diamond. Of course, he was the instrumental guy in leading our Krumovgrad project, and we're very proud to say that that project, which is now ramping up, is one of the few examples of a mining project which has come in under budget and pretty much on schedule. Really, the ramp-up itself is really looking like it'll come on pretty much online with what we expected. I could say you could probably count them on one hand how many of the mining projects around the world do that, and I could tell you that John is the guy you want running your projects area.
He's the guy that I call him the old Luddite when it comes to the style he has, he is a very strong leader and a great guy for helping out the company, too. He's not only a good project guy, but he's a good team member in terms of sometimes putting the rest of the team back into the right conversations on the ground, and making us think hard about what we're doing. Anyway, John, thank you very much for that. I think we all know about forward-looking statements. I want to go back here. Hope I know how to do this. Oh, there we go. I just got trained today, and I'll have to remind you, I'm the Luddite of the group. Even though I talk technology and innovation, I can't do it. Okay. Oh, this is a repeat. Sorry.
Maybe I should back down again. Wrong one. Stop. We'll run through a few graphs here. Right?
Yeah.
50,000 tons of copper concentrate. In the first quarter, we were in terms of earnings, adjusted EBITDA, the fourth will be going up. In terms of maintaining costs, last year we were CAD 659. This year we're basically dividing between CAD 659 and CAD 640 in terms of maintaining CapEx spend. As the Krumovgrad completes, we'll drop back to just maintaining sustainable spending, maintaining CapEx spend. Last year it was a bit, this year it'll drop. Share price performance. We look at relative to all the indices, you can see that it's been going up quite a bit well before we indices. We look at the kind of the annual divide here.
Well, the company vision is really to be a progressive gold mining company, innovative. All those things are gathered into that foundation innovating and we're trying to basically understand what makes a better day. All right. In terms of the What we'll do, we'll have a look at a number of different things that we've been busy with. If you don't mind, I'm just going to walk over here. It's going to be basically a prompt for me to bring up a number of things that I want to show you of some of the innovations that we're doing. I'm going to start off by looking at our MineRP activity. If we look at MineRP, which is primarily focused on the activities around planning.
I'm just going to run through this and I'll pause at different points so that you can understand what it is that we're trying to seek in terms of opportunity with this. It's primarily focused about rethinking our planning, scheduling, and execution, and it's really built on the fundamentals of moving from strategy through to operational and execution. The platform really is looking at a number of different elements through a spatial database that allows us to get a certain reporting and control and analysis. Now, if I just stop this for a moment, the key things that we're working on at the moment as part of our rollout of MineRP are broken into the design, basically the blasting activity and the design of the blasting elements, and then from there on to production scheduling.
What we're doing is we're linking to a resource database, and then we're linking that down to the operation of the equipment, and then service functions which are important to this particular activity, such as survey and reconciliation. The key thing to all of this is the ability to take these elements, link those to a spatial database where you have an XYZ coordinate plus metadata such as density, tonnages, grades, response to flotation or cyanidation, whatever that might be, and then be able to link this to all of the different softwares associated. We've been working on this with MineRP actively, and we're now at the point where we're ready to actually introduce these three elements.
We're busy at the moment with user acceptance testing, and we will within the next couple of weeks actually be running this on a parallel process with our existing planning system. I'm just going to move this along a little bit. What I'm going to do is I'm going to run a couple of videos in the background here and talk around these. My intent is not to actually talk to them, but it's just showing you different things that you can actually do within the software. What you see here is the infrastructure of the mine, and you can see, in this case, elements of the block model.
What it allows you to do is to import that information from your resource database and then plan that, whether it is at a high level or a low level in terms of timing. I will come back to that in a moment. This is traditionally done in mining operations where you have these disparate files which are not connected. Anything that you do moving from the resource database into your planning disconnects all of these elements, and it means that for the moment you start working on that overall resource, you are actually disconnected from everything else that you are doing. Just pause this for a moment. The key things about this are this ability to plan across different horizons. We have four different planning horizons, different connections to databases. Every time something was basically done, you lost that connectivity.
Basically, what we can do is we can actually control all these planning horizons, single database, fully linked. Any changes you can basically bring back in with no other steps. The key bit as a next step is this piece in the bottom. What this is are the different elements of the planning that we do, and this is referred to as parametric planning. Once you have this thing designed and functional, you can then replicate what are currently manual processes which take a lot of time. You have a person planning each and every sequential activity that takes place, whether that is advancing on a blast or the blast design itself, or the drift or the development or so on. What you can do, the bulk of these are actually the same.
What you can do is you can automate that so that you only work by exception. What difference does that make? Instead of taking many hours or days to be able to make an adjustment, you can do that in minutes. The importance of that is you are able to optimize your plan at any given point, something we really cannot do at the moment. At the moment, we are working on the planning, blasting activities and working on the planning through to scheduling. The parametric planning is really going to give us the power with the next phase of this activity. If we move this thing forward, what I want to talk to you about here is what happens with the blasting design, for instance.
What you are looking at here is you are looking at the design and the layout of all the infrastructure. Here in this piece is your design now to actually remove the ore. You are accessing, drilling from the top, charging that with explosives. You are actually basically retreating at 2-meter intervals. What this is this piece of software that is actually linked through to MineRP, which allows you to do the design. Actually, these green tubes here are the explosion energy in spheres through that level. You can actually basically predict what is going to happen with the blasting as you design it. It has been necessary to change to this to really get this seamless interaction with MineRP in terms of what we do going forward.
The next piece is you can look at this from a blasting design point of view. What you'll see is that you're able to break this thing down into a number of hole-by-hole developments that you can see here. Here's your access drift. This is where you're going to remove the material from the bottom. This thing is going to be moving backwards into the page. What you're looking at here is each one of the holes as you retreat. The next step is that what we do is we bring up the integrated design, which takes account of the resource through the plan, we start to move towards the schedule. Here what you're going to see running is basically the development of the mine and the access, the removal of the material, the treatment of the material.
This first piece is really about the sort of higher planning horizons, we're going to move down to the scheduling process. You've got many horizons in one plan, schedule optimization at every level, you've got this three-dimensional reference. What that does when you have all of this running is you're able to actually optimize by running parallel three-dimensional optimizations, as you can see here. This is now moving to an optimization process, which really isn't possible for us to do at the moment. As we progress with this, we compare different options, we move to short-term scheduling. We've gone long-term, now we're going short-term, we're looking at activity-by-activity, location-by-location scheduling down to a shift and task level.
When you actually execute that in MineRP, we transport this from MineRP into the control application, which is a Sandvik as opposed to MineRP application, as you can see here. We're interfacing with what's called OptiMine. This allows us to take that information in a schedule for different activities. For instance, I don't know if you can see this, these are drilling activities, blasting activities. You've got maintenance. You've got all sorts of different things, roof bolting and so on. This is basically developed for the person who's monitoring the execution. They can see the advance in real time against what is actually happening.
What they can do is if anything needs to change, they can change the schedule. That gets driven down to the tablets on the equipment and to the people who are operating, they're able to track what is going on. Basically what we've done is we've moved from this integrated design and a high-level planning down to short-term scheduling. That basically completes this part of the presentation. What I'll do is I'll stop at this point, we'll move this on. The next thing is all of this activity is really perfect for us to interface with this other digital activity that we've been working on, which has been the development of a smart center. What this is something that iron and steel, the big companies, BHP, Rio, have been doing for many years now.
It's unusual for a company of our size, but we believe there is opportunity for us to actually bring together the different time horizons and planning and activity, scheduling and execution activities with people working together from a broad range of disciplines. For instance, we have mining people, mill people, we have technical services, mine technical services, process technical services, and all of those were previously disconnected. With some of these tools like MineRP, this is the opportunity to take advantage of that. We started this in September last year. We now have the mine and mill in this particular room. It's already generating benefits. The next stage is that we'll introduce mine technical services and maintenance, and then we'll have the concept of how we roll this out and how we create value.
This is something we're going to be progressing through the course of the year. The next piece is what I touched on a little earlier when I talked about the planning and the execution, the bringing back the information. There are various tools to do that, but the tools typically have been, let's say, a little crude relative to where the current state of technology is. One of the things that we've been doing is we've been working with drone technology, an example of which you can see here, working with our partners, Exyn. What this is doing is it's allowing us to actually fly into stopes because this is an autonomous drone. It doesn't need somebody who's actually piloting it.
What happens is that you create the target points that you want it to go to, and the drone will follow that pattern and map what's going on in the stope. Why is that important? Why would that add value? Sorry, I just want to resize this a little bit. There we go. Why is that important? What we're finding by doing this, it's not just the level of resolution, but we're seeing things that we previously couldn't see with the tools that we had. What you're seeing now is you're seeing the outcome of one of those surveys, and the drone has flown back into the stope from the top, and it's now going to exit at the bottom of the stope from which we remove the ore using wirelessly controlled equipment as opposed to having people in these things.
What I'm going to show you next is I'm going to show you what this looks like on a three-dimensional basis, what it's created, and then I'll show you some comparisons in a moment. The drone returns. You can have a look at the equipment, which is behind me here to the right, if you want to see that in a little bit more detail. What happens is you get a three-dimensional scan of the stope, as you can see here. We have a couple of 3D-printed models if anybody would be interested in seeing that. This is the access from which we would actually do the drilling, and then we would do the blasting and advance this backwards. What we do is we extract from the bottom.
This allows us to go in and see what that looks like at any given point. That's something that we couldn't readily do previously. If you have a look at, for instance, this was the historical method on the left-hand side, and what you see is 50,000 points. This would take, typically we'd say four hours, but we're being generous, between one and three hours to do this. What would happen is you put a device in through the top, and you would scan to the limits of what was available. Whereas when you fly this thing with the drone, you get 35 million points relative to 50,000, and it takes two minutes for data collection and about eight minutes of post-processing. The really interesting thing is if you do that scan. We can live with that, right?
Look at it on the right-hand side here. These are the two scans overlain. Actually, what you're seeing here is what could not be seen by this previous technology. Why is that important? It allows us to do a number of different things. The most expensive consumable you have underground is blasting materials. What this is telling you is you've actually taken out more material than you expected. If you just go to drill that based on this profile, you're going to be wasting explosives. It gives an opportunity for us to have a direct impact on costs, a direct impact on things like ventilation and other activities. It's quite marked in terms of what it can do. In terms of the smelter, Rick's already mentioned that we've had a good couple of years.
We've seen the new team at the smelter that's been in place now for about three years, really starting to find its stride, making a significant difference. As the capital cost spend has been going down, we've seen the performance of the smelter improve. We've talked about some of these things before, I won't spend a lot of time on it. One of the biggest impacts right now is our primary smelting device called an Ausmelt, which smelts around 750 tons a day. That's been basically through the years in the different campaigns. These are the different relines. If you look at the tonnages we've managed to put through, it's been sort of capped at about 150 to 180,000 tons.
We've identified ways of being able to increase that, but more importantly, we can actually measure it, which is something that's unique to this particular vessel for the industry. Nobody else does this. What we can do is we can actually measure what the consumption of the lining is and project that forward. This is why we're confident that we're going to be moving from as low as six months up to 18 months initially and 24 months. What that does for every reline that you save, you save CAD 5 million worth of direct costs, and you add about CAD 14 million worth of additional revenue. It's at the point now with the reline that we plan at the end of this year, we will have no relines in 2020.
Where in this model, we would have had at least one and possibly two based on the old design. That's one of the reasons why we've had big improvements. You'll notice if you read this, there are other things, better power management, increased oxygen enrichment. Some of these things are unique to this smelter of this design. The other thing that we've been doing is we've been utilizing better the equipment. It's not just the Ausmelt. If you have a look at the converter, this is the secondary vessel. It's a batch vessel. There are two of these. We installed them at a cost of CAD 40 million along with the acid plant three years ago. We were recognizing that every time we had to rebuild, there was an issue with our capability to treat the material that was coming from our primary smelting vessel.
At the time, what we did was we identified a CAD 1 million spend to take an old converter and to make that a holding vessel to create additional capacity. If I can just sort of at a high level tell you what's going on here, you'll see there's three different time horizons. Actually four, sorry. If you look at these trends, what was happening initially, we were getting sort of 600 to 700 tons a day during the time we were doing a reline. Now, bringing in the matte holding furnace, that gave us an opportunity to store some material. It also allowed us to treat at a higher rate and provide some buffer. What happened was we moved this trend so that we were actually increasing the tonnage per day treated.
If you go back to October 2016, we were operating somewhere between 600 and 630 tons, and we're now running consistently around at 780 tons. That is north of a CAD 20 million NPV, which means basically a one-quarter payback. We avoided the cost of CAD 20 million for additional converter. I'm just going to show you one other item now that we're busy working on. I've talked about a number of different things that are more production, productivity related, but this one is more safety related. If I can just share this with you. This is a virtual reality simulation. It is at the Ausmelt, and it's actually taken from images of the operation. If I start this thing up. Sorry, it looks like I didn't bring up the actual video. Try that again, shall we? All right.
What you're looking at now is you're looking at the. Just bring this one up. You're looking at the video of the person taking hold of the virtual control. Sorry, that's on the right one. What you see here is the mud gun rotating around the body of the vessel. This is a five-meter diameter cylinder, 10 meters high. At any given time, it's got about 60 to 90 tons of material at 1,200 degrees C. Getting things right in this environment is really important from a safety point of view. Basically, we were training people on this device in the field. The problem with that basically was that if you move this thing out of the way, you'll see that it's a fairly confined space. Here, you've got a launder, a little space between the launders.
All of this is hot at any given time. Basically, understanding what was going on, managing the risk associated was important. Here's the mud gun coming up and coming across. What this does, it allows you to drill the tap holes and subsequently close them after tapping. I'm just going to move that head a little bit. Here's one of the tap holes. This is a slack tap hole. If you drill a hole through this 500 millimeters, you're going to get material coming out about 20 tons in, say, five minutes. What happens is the mud gun puts pressure on the hole and closes it.
This virtual reality simulation where a person is basically remote, and he's able to simulate with the actual control panel what it looks like, exactly how he would go about operating the gun, and also hazards that are in the environment. If you look over here, you can see there's actually a water leak simulated, and that shows then there are consequences for mistakes if you haven't done your inspections. Opening the hole now as he is doing with the drill, what's going to happen is you're going to get an explosion when the material comes out. You can actually train people in a remote environment without putting them in harm's way and get them to understand the sort of risk they're facing. As you can see, you can actually simulate all sorts of things.
What this is doing is it's showing the material tapping out here from this upper layer. This is the matte containing the copper. This is the slag, and here's the inside of the furnace. In terms of your ability to visualize things, this is a dramatic improvement. Now, this is a test for us. It's working extremely well, and we're now looking at how can we use this elsewhere. We could use this for all sorts of training, but also productivity and operational improvements. I think that concludes what I was going to talk about. See, Dave gets to do the fun stuff. That's okay. Okay. Okay, I'm going to talk a little bit about growth. The first thing is our Krumovgrad. Rather than me speak about Krumovgrad, we put a short video together for you to explain it.
I just wanted to bring you to speed. The Krumovgrad project is currently on the ramp-up phase. Construction is now completed. At the end of March, we were essentially complete the construction phase, so we're in the commissioning and ramp-up phase. Everything actually is going quite well, and we anticipate that we will be in commercial production in this quarter. This has been a great success story, and I can just put the video on now and make it easier to explain the whole piece. Oops. Did I lose it? Oh, it's beginning. All right, here we go. Okay. Oh, it's down too low again. Thanks, Dave.
Dundee Precious Metals is a Canadian-based international gold mining company with operating assets including the Chelopech Mine and Krumovgrad project in Bulgaria, the Tsumeb specialty smelter in Namibia, and the development stage Timok Gold project located in Serbia.
We achieved first gold production at Krumovgrad in the first quarter of 2019. This is a very significant milestone for the company, and that we'll be transitioning to a significant amount of free cash flow growth from the Krumovgrad project, and as well will put us into a mid-tier classification as a gold producer.
As the first mine to be built in the Balkans in 40 years, it's a pivotal milestone for the company, the community, the country, and the region. The property, located about three kilometers south of the town of Krumovgrad in southeastern Bulgaria, consists of the Ada Tepe deposit, which contains 800,000 ounces of gold reserves. It is a high-grade open-pit deposit with reserve grades of four grams per ton and a very low strip ratio. The deposit consists of the wall zone and the upper zone and will be mined in four phases. Krumovgrad's relatively simple processing flow sheet encompasses crushing and grinding of the run of mine ore, followed by flotation to produce a high-grade gold-bearing concentrate, which will be sold to global smelters for further processing.
Also, tailings from the plant are thickened and disposed of into an integrated mine waste facility, a design which uses mine waste rock to construct cells for the facility. Innovation plays a central role to all of DPM's operations. At Krumovgrad, technology is also being used to ensure the quality and efficiency of the project. This includes a blast monitoring system to ensure more accurate ore classification and grade control. By being persistent, flexible, and innovative, the company overcame many challenges that the project encountered since the original feasibility study was completed in 2005.
We made a number of changes to address the concerns of the local community. First, we eliminated use of the wet tailings facility. We also changed the flow sheet to change the technology we use for that to address some concerns they had around the flow sheet. We also shrunk the footprint by 50%.
The company is proud to say Krumovgrad is the first project to be permitted in the European Natura 2000 ecological network. After receiving construction funding from the European Bank for Reconstruction and Development, the project has been designed and developed to meet their environmental and social performance requirements. The project will contribute significantly to local communities by employing approximately 230 people and contributing funds to education, infrastructure, health, culture, and small business development projects. The project could already be considered a success on two measures. The project has come in under budget at CAD 165 million compared to the original budget of CAD 178 million and recorded 2.6 million work hours without a lost time injury.
The mine will produce 100,000 ounces a year over the first five years of its life. Over the eight-year life of the mine, it will produce 85,000 ounces a year on average at a cash cost of CAD 400 an ounce. We're exploring around the mine, and we've discovered a number of satellite deposits which we expect will extend the life of the mine as well.
The project will significantly increase Dundee Precious Metals' consolidated annual gold production to approximately 350,000 gold equivalent ounces, and also significantly increase its cash flow, unlocking significant value for the company, its investors, and local communities.
I think the key message here is that we are finally at, after 2005, almost 15 years of hard work, bringing this project to fruition. A lot of effort, a lot of the employees that have worked in the company to make this happen. It's been a great success. We'll listen to that. Our next growth potential project is called Timok in Serbia. Just talk a little bit about it. You can see Timok project here in Bor, the region in Bor, near the Timok Copper Project. You can see essentially the exploration license here called Potaj Čuka and Umka. Those are the license areas. Bor is located here. The Č ukaru Peki, the Timok Copper Project that Nevsun or now Zijin owns is located right here.
This is a zoom-in of the license area and the three deposits, the newest discovery, Korkan West here, and then the two other discoveries, Korkan and Bigar Hill, and to the south, Kraku Pešter . The four deposits themselves make up the Timok Gold Project. Let me zoom in on just this area here. These are really the core ones that we are doing the studies on now. The original test work that was done indicated that a large component of this deposit was oxides, and we produced a new resource estimate late last year that basically found essentially 2 million ounces of oxides and sulfides, about 1.1 million ounces of oxides, 900,000 ounces of sulfides in these four deposits shown here.
The work that started on that was the leach test that we did earlier this year or late last year, I mean, which indicated that we were getting very good recoveries on the oxides. We got over 94% recovery on the oxides in these two ore bodies, which are the two largest ones. We got about 70% on the Korkan West deposit. If we just zoom in on this and we can show you in three dimensions here. This is the drilling. You can see that it's fairly well drilled already in these two bigger deposits, Bigar Hill, Korkan, and the other one, Korkan West. The red represents the oxide material, which is sitting at the top, and then the yellow represents the transitional material, and then the blue represents the sulfides underneath. Essentially a pretty straightforward deposit.
This is just running in the background. Sorry. Is this a PowerPoint? Okay. I did want to touch just briefly on the environment, social, and governance, which is really a key part of any mining company's core success factor now. We think we've excelled fairly well in this area. We have a very clear corporate responsibility policy, and that ties together really our focus around health, safety, environment, and social. We've probably integrated, I'll say that, whole corporate responsibility into one policy. That's a bit unique, I think, in the industry. Many are still focusing on individual policies for each area. We've tried to integrate ours into one corporate responsibility policy. Just to give you some indication of some of the ESG performance that's occurred in the company, really over the last four or five years.
As you can see that we've had significant growth in production, both in terms of production of copper concentrate and production of blister copper. We've also at the same time improved energy intensity, reduced Chelopech's energy intensity by 13%. We've reduced Tsumeb's energy intensity by 50%. Water intensity, we have seen a slight increase in water intensity at Chelopech over the last little while, but we expect that to turn around and start to come back down again. The Tsumeb water intensity has improved by over 42%. Greenhouse gas emissions, the emissions intensity has dropped for Chelopech by 27%, by Tsumeb by 41%. The reduction in the SO2 emissions, which is also a greenhouse gas, was reduced by over 89%. You can see our totalable reportable injury frequencies have been reduced by 85%.
In terms of diversity, if you look at our organization, currently have 21 out of our 45 senior managers are women. That's quite unheard of in the mining industry. That's almost 50% of our senior managers in the organization are women. Then community investment, we spent over CAD 2.2 million invest 2.2% or million? Million. It says percent million. I know which one that is. Which is it?
Million.
Why is the percent there? Mistake. CAD 2.2 million invested in 2018 on community investment. Is that just for That's for all three sites. Yeah. Just an indication that we are working very hard on this, we do believe that the demand on the mining industry is to do more than just comply with the regulatory requirements and more than just get social license. It's really about having a social purpose, about driving that social purpose for social upliftment in the communities that we work in. We do really treat our community partners as partners and work closely with them to decide what are the key, I'll say, concern issues they have, both about our site and what we do, but also what are their key concerns in terms of social upliftment in the communities, whether that's education, healthcare or poverty, drinking water, et cetera.
We try to focus our attention on those areas. Just as a way to kind of put some perspective to this, as you know, there's more and more indices that are being used to score companies in terms of their ESG performance. This is just one of those indices. It's called the Arabesque S-Ray Index, I think it incorporates 7,000 of the world's largest corporations. I don't know why we're in the world's largest, but we're in there, in one of those 7,000. You can see on the two axes here, one is the global compact score, which is the vertical axis, which is really the United Nations Global Compact on human rights, labor rights, the environment, and anti-corruption. That's the score on the left, on the vertical axis. On the horizontal axis is the ESG score, the environmental, social, and governance score.
You can see where Dundee Precious Metals stands relative to those 7,000 companies, very high on both scales, on both ends of this spectrum. We're already positioned fairly well as a company in the ESG world, but we know that there's still room to improve, so we're continuing to focus on improving our performance in this area. I think that's the end of that. Just in summary, we are a growing low-cost producer, producing 350,000 gold equivalent ounces at an all-in sustaining cost of between CAD 650 and CAD 730 by 2020 and beyond. We'll be a low gold cost producer in the lowest quartile of the industry.
You can see that in terms of EBITDA growth with Krumovgrad, we'd be making a total of CAD 194 million once Krumovgrad comes online, in terms of EBITDA, less sustaining capital, and at an enterprise value less sustaining of about three times our multiple on that. In terms of our current indices, I'm going to keep running, but in terms of our current multiples, we're still trading, relatively speaking, at a low multiple to our value. I'll just stop this at this right point here. You can see here this growth of our earnings that will come from Krumovgrad. Then if I just let it play a little longer here, that very quickly jumps away, but that's the multiples that we're trading at. Okay, I think enough of that. The last, just wanted to talk briefly about our disciplined capital allocation.
Dundee Precious Metals is a Canadian-based international gold mining company.
What the heck is that? Okay. There we go. This is what I wanted. Obviously with heading into free cash flow growth in the company, the key question we're getting asked by many shareholders is really how do we look at the capital allocation? We spent some time putting together a disciplined capital allocation framework, which is shown here. Really it's quite simple. In terms of this framework, it's all about maintaining a strong balance sheet. Obviously, priorities being around reducing or eliminating any debt and building a strong financial or a cash position. Once we're in that stage, which I think we're headed for, really the decision comes around reinvestment in the business or return of capital to shareholders. We don't see those as mutually exclusive. We see those as decisions that are made by the management and board in terms of how that happens.
Essentially, we can see that with the kind of growth that we're going to get in free cash flow, we can actually reinvest in the business as well as return some capital to shareholders. That's really the strategy going forward. The board will discuss once we get to a point where we're reaching that level of sort of a strong cash position and growing cash position, that we would make those decisions about how best to do that, how best to return the capital, whether that's through dividends or share buyback, and then how much of that money will go into reinvestment in the business. Essentially, that's the simple framework. Okay. Okay, that's I think essentially the end of the presentation. I know it was a bit long.
Just want to thank everybody for spending the time, and, as I say, I'm the Luddite of the group, so I'm still learning how to use these tools. Thank you for your attendance, and please, at any time, come up and talk to any of us, and we can put some of these things back on the screen if you want, and we can talk about them. We do have, by the way, I just wanted to introduce Exyn, who is the drone Not the drone manufacturer. They're the software guys who made the autonomous drones. They're here today. If you really had a question about how to fly this in your backyard or something, or you wanted to fly it through your house and get a survey, talk to Nader, He'll tell you how to do it. Okay, thank you.