Geology & GeoMetallurgy

Geology & GeoMetallurgy

  • To participate in the 911Metallurgist Forums, be sure to JOINLOGIN
  • Use Add New Topic to ask a New Question/Discussion about Geology and Geometallurgy.
  • OR Select a Topic that Interests you.
  • Use Add Reply = to Reply/Participate in a Topic/Discussion (most frequent).
    Using Add Reply allows you to Attach Images or PDF files and provide a more complete input.
  • Use Add Comment = to comment on someone else’s Reply in an already active Topic/Discussion.

Geometallurgy and mine planning (10 replies)

(unknown)
8 years ago
(unknown) 8 years ago

Does anyone have experience and success incorporating their geometallurgy models into mine planning and pit optimization processes?

JohnnyD
8 years ago
JohnnyD 8 years ago

Absolutely! Any information that can differentiate the behavior of material as it passes through the system in terms of throughput/recovery/cost/product quality creates an opportunity. Where the characteristics of different zones in the ore-body are known, this can influence pit and phase optimization (or stopes in underground mines). Where the different characteristics are interspersed in the ore body, there is a an ore control/cut-off/stockpiling opportunity. It takes a significant shift in organizational thinking to exploit these opportunities and a higher level of cooperation between geologists, miners and metallurgists. In addition to the change in mine planning practices, it takes a different processing/metallurgical mentality: stop testing composites - we need to know how the components behave individually and together, stop bending out variation - exploit it, design headroom into the system so the deliberate variation can be accommodated, forget the concept of "Steady State" operation - that is ok day to day but not over the life of mine. If you go down this track you need to also implement Activity Based Costing and Theory of Constraints - but you will never look back.

(unknown)
8 years ago
(unknown) 8 years ago

It helps but the key is incorporating the physical mining activity with what it computed as a model. The computation process is the core of geomet model like any modelling study. How it is done, how realistic, is it mechanistic? If the evaluation is positive then mining activity can be guided and the benefits are going to be promising.

(unknown)
8 years ago
(unknown) 8 years ago

My vision is to incorporate a geomet model into the mining model itself.. one model, to work with, to optimise throughput and recovery, maximising value. It is very much a "team" model requiring input, analysis, review, and adaption from all stakeholders. The variables that make up a geomet model are varied and start with the basics, the geology, the grades, but then one starts adding the expectations of what the dirt can yield and the impacts that characteristics might/will possibly in materials handling, comminution, and ultimately recovery. It needs very much to be a working model, evolving as actual data is realised and reconciliation is performed. Seeking to understand the performance through a plant helps both the short-term, as well as the long-term. Long-term strategies should incorporate these qualities in a way that maximises the value from a deposit. I have but seen steps in the right direction towards this vision, and I have seen a fair few deposits mined >20, but I know adopting it can only serve to add value to any long-lived deposit.

(unknown)
8 years ago
(unknown) 8 years ago

Thank you for your in put gentlemen!

I do tend to agree that we only have taken small steps of converting geomet model information in to our mine planning and scheduling packages. We normally use simplified differentials to transfer the information where the bulk of the information remains in the geomet model and is not utalised in scheduling.

Thanks again for your contributions

(unknown)
8 years ago
(unknown) 8 years ago

You probably realize from what everyone is saying that this is an area requiring full comprehension. A lot of champions of geomet do not understand it themselves and understandably so.

Serious work is now going into changing this misnomer-GEOMET: geologists yes, metallurgists yes, but everyone else is then left out. Miners have to be persuaded to be part of geomet, so are other resource players simply because the name excludes them. Geomet or whatever the new name is going to be is everything as long there’s traceable VALUE all the way.

The traditional mining systems are designed for rigidity at both mining and processing stages. We have these large shovels or our plant can only take in this kind of ore and hence the general tradition to homogenize and ‘steady-state’ everything every time! Successful resources industries have been created without ‘geomet’ so unless there’s traceable tangible VALUE, planners will not work with such models. A lot of VALUE is embedded in knowing in advance critical areas like drill-ability, blastability, loadability, crushability, grindability, recoverability, energy and water usage, etc. One then needs ability to define these variations at source enough to predict outcomes in VALUE. Good luck in modeling non additives, that remain moving goalposts..

(unknown)
8 years ago
(unknown) 8 years ago

You can collect geomet data as much as you want , you can have the best model possible on hand, at the end of the day what really matter is if this information would be used to mine and mill more effectively. This require a lot of mentality change in operation, unfortunately it seems to me we are not yet there . It has been my experience that geomet modeling has been used only when 1) things at the operation are not going well and 2) you can prove the geomet data would be a game changer. In my opinion,it would take time for geomet modeling to be fully integrated in the mining plan and help maximize value; yet, as long time practitioner , I say let keep going, one step at the time .

Helena Russell
8 years ago
Helena Russell 8 years ago

I have been working with some scientists and computer engineers in a collaborative project seeking to build integrated workflows from geology through to scheduling and it is encouraging to hear there is an ongoing need for this. Our focus is research, which is often limited in its industry impact because it only addresses a narrow part of the value chain. A broader challenge in this regard is the ability to provide transparent and testable methods, ie through public domin publication: is there a way real data from resource drilling (possibly mined out) could be made available for this purpose do you think?

Maya Rothman
8 years ago
Maya Rothman 8 years ago

The best published examples in operating sites are Batu Hijau, as Fatih mentioned, and Escondida. These have both got geometallurgical information into the medium term plan but the question comes in at what data resolution can be maintained before it gets unmanageable. There has been some great work done over the last decade in development of geometallurgical tests and improvements in geostatistics but these are yet to be used at any site in a real 'game changing' way. The work I have done in the area has shown that the soft issues, such as integration of disciplines in an operation, are just as important as the technical side of things and any project requires buy in from the geological, mining and processing teams, as well as support from the corporate office. I've been working on development of a systems engineering approach to try and address this and have been having good success at one site in working from the process end to integrate day-to-day process mineralogy data, which will trickle feed back to the mining model.

(unknown)
8 years ago
(unknown) 8 years ago

The biggest plus related to integrating ore body knowledge (in whatever sense) in a single model is that it breaks down the silos around the disciplines, many of us, are multi-disciplinary and can readily see the value of working with an integrated model through mine planning and execution. It helps level the playing field, so we all can readily discuss elements that are significant for each of the disciplines. Anything that takes into account knowledge already paid for, and makes it available, easier to reconcile and work with, makes better use of money already in the bank as long as a company sees the value in trying to actively manage extraction (rather that just sitting back on their heels or just satisfied with fighting fires as they rage).

As most companies mine more than one ore body (excluding OD? or is it actually made up of multiple ore bodies aka zones, we know so.), the learning and system improvements from giving it a go at one deposit or mine, can be readily applied to the others in a never-ending improvement process. I would rather engage the ore body with my eyes and ears open, and respond to its breathing, rather than think the original estimate was all that is required. It really is just a huge information loop. Metallurgical expectations are often incorporated into models in the form of predicted products or NSR calcs, but the levers which result in those predictions or the other factors that might influence realising those predictions are not readily there, or conceived in a way to be related to the planned end. There is no one size that fits all in this game, but with each attempt, we inch forward towards the vision.

Bob Mathias
8 years ago
Bob Mathias 8 years ago

It is not clear from your question and comment whether you are more interested in the technical or change management challenges of interacting with planning and scheduling, so I will add a few comments on both aspects. Many of these will mirror the excellent comments and recommendations already provided in this discussion.

I have been involved in the development of numerous geometallurgical models over the years, predominately for copper mill and leach operations and projects. In most cases, the results of the models were intended as inputs into mine planning and valuation models, while in some cases the results were more focused on inputs to short term metallurgical blending plans. One of the more successful efforts was my role in supporting development of the Escondida geometallurgical models during 1999 – 2004. Leo Flores provides a fairly detailed view on development of the mill throughput model in a paper that was given at the 2004 Canadian Mineral Processors meeting, and is available as an SGS Technical report. This report gives some insight into the technical approach taken for sampling, estimation, metallurgical modeling, and reconciliation that may be of interest to you.

The Escondida geometallurgical model was formulated as a planning project and not as a “geomet” project, with the objective to provide more accurate estimates of the key metallurgical value drivers for mine planning, valuation, and project design criteria. There were two main components to successful implementation of the project into the planning process. The first is recognition that key metallurgical parameters in hydrothermal deposits are derived from spatially-variable ore characteristics, thus causing metallurgical variability through time in a mine schedule. The second is creating a consensus from the operations leadership team of performance criteria on the level of accuracy required for metallurgical forecasting in budgets and medium-term plans.

I assume that the current mine schedule would have estimates on the key metallurgical value drivers (recovery, production rate, concentrate quality, reagent consumption, etc.).
If the current basis for metallurgical forecasting can provide expected accuracy for metallurgical forecasts in the mine schedule, it would be difficult to make a business case that anything different is necessary. But, if existing data suggests that orebody variability is greater than the performance criteria for metallurgical forecast accuracy, a business case for changing the basis of the forecasts should exist. Hope this helps.

Please join and login to participate and leave a comment.