Schedules, assumptions and every run are kept against your account.
At least 8 characters.
This server is invite-only. Ask whoever runs it for the code.
A project holds a block model, its assumptions, and the history of every schedule run against it.
Datasets available on this server.
A zip of CSVs. The quickest way to get the shape right is to download one of the datasets above, open it in a spreadsheet, replace the contents and send it back. Anything wrong is reported by file, row and column before a single block is scheduled.
A licence belongs to an organisation, not a person. Everyone in one shares its projects; the seat limit is what you are selling.
What stops one customer taking the whole machine.
0 for unlimited.
Issues a one-time password and ends that person's sessions. They cannot do anything until they replace it. Hand it over yourself — nothing is emailed.
Set this to whatever you already use elsewhere, if you would rather not carry another one. At least 8 characters. Your other sessions are signed out.
A snapshot is every account, project and run in one file. Taken nightly and kept on the separate storage volume. Download one to hold a copy somewhere that is not this server.
Block centroids coloured by grade, with the drilling. Drag to rotate; hold Shift and scroll to zoom. Nothing is scheduled here — this is the model the assumptions below will be applied to. Open the full view for bench plans and the summary figures.
Build a block model from the drillholes and look at it before anything is mined, costed or scheduled. You get the deposit in three dimensions, the variogram drawn over the points it was fitted to, the validation checks, swath plots and a grade–tonnage curve. It takes seconds to change the search and produce another, so this is meant to be argued with rather than accepted.
It declares nothing. Every block comes out
UNCLASSIFIED and the estimator's opinion sits in proposed_class,
which a Competent Person has to move across deliberately.
Where stage 1 asked where the metal is, this asks what a shovel could dig. The estimate is combined into the selective mining unit — the smallest volume the equipment can actually send one way rather than another — the ultimate pit is cut from it, and the pushbacks and bench plans come out of that. You get the pit in three dimensions, a section through it, nine bench plans, the pushback table and the stripping bench by bench.
It is not an Ore Reserve, and the pit is not a mine design. There are no ramps in it, no minimum mining width, no geotechnical domains and no operational judgement. It is where a mining engineer starts.
Set mining_block_x, _y and _z
below to the size of that mining unit. Left at zero, the mine is planned on the
estimation block itself, which claims a selectivity the equipment usually does
not have.
Every cost in this software has been a flat rate typed into economics.csv that nothing ever checked. This estimates what a mine of this size, strip ratio and region ought to cost to build and to run — and, more usefully, places the rates the schedule has been running on all along inside that distribution, so you can see where they sit.
It is an AACE Class 5 screening estimate, accuracy −50% to +100%, and nothing here feeds the schedule. The economics the schedule runs on are unchanged. It must not be used for an investment decision, and a single figure taken out of it is a misuse of it — the range is the answer.
Mining method matters as much as commodity: the open-pit and underground models for the same metal differ by two to four times, so this is not guessed for you. Set the region, the orebody quality factor and the throughput below.
Every net present value this software has produced has been a pre-tax one, and on a mining project the difference is routinely a third of the number. This runs the royalty, the capital allowance pool, the loss carry-forward, the income tax, the withholding and the indirect taxes over an indicative life-of-mine cash flow — in the order they are actually charged — and then screens every other jurisdiction in the pack on the same cash flow so you can see where this one sits.
It is not tax advice and it is not a filing position. The rates are a screening compilation of each country's general law at a stated date. Stability agreements and treaty relief routinely override that and neither is modelled. Verify every rate against the statute before it informs a decision.
The cash flow is built from the cost model, so its accuracy is the cost model's: AACE Class 5, −50% to +100%. Timing is what a fiscal comparison is most sensitive to, so the build period, ramp-up and dividend payout are settings below rather than assumptions buried in the code.
Schedules the pit stage 2 laid out, builds the cash flow that schedule produces, and runs it through the fiscal regime stage 4 chose. The result is a post-tax net present value — the first this software has produced. Every figure before it was pre-tax, which ranks two deposits in the wrong order whenever they sit in different countries and says nothing about when the owner sees money.
The costs are the project's own from economics.csv, not the stage 3 benchmark: a Class 5 screening estimate has no business being the basis of a valuation. It also runs the same cash flow through every jurisdiction in the pack, so you can see where the deposit is worth most.
This one takes minutes rather than seconds: it solves the
schedule. The solver budget is time_limit_s below, and a
schedule that stopped on the clock is a lower bound on what the deposit can
do rather than an upper one — the page says which happened.
Every choice below has a default already selected. The defaults are the agreed assumptions, and each alternative is offered only where it is genuinely defensible.
Grouped by what they affect. Anything marked fixed is not offered as a choice, and says why.
The full report: schedule, cash flow, aggregation, the solver's own account of itself, and every assumption behind it. The five stages above are the model built one visible step at a time; a run is all of it at once.
This software does not declare a Mineral Resource or a Mineral Reserve. It records that a named person has taken personal responsibility, and pins that record to the exact result they saw — the input hash, the calculation version and the value. A signature cannot later be moved onto a different answer, and it is withdrawn by recording a withdrawal, never by deleting it.