Block model and constraints in, schedule out

A long-horizon model is normally trimmed back to stay solvable in reasonable time. The trimming happens because the solver runs out of capacity, and for no other reason. Cognomine is built to hold more of it in a single solve, including:

A schedule that solves fast is not the same as a schedule you can hand to a crew. Most fast schedulers buy their speed by dropping the constraints that make a plan executable, and the planner spends the next week putting them back by hand.

Cognomine holds those constraints in the solve. Drill-and-blast lag is modelled, so the schedule does not ask you to dig ore that has not been shot. Digger separation is enforced, so the plan does not scatter equipment across five faces to chase grade. Slope constraints are solved in, not applied afterwards and reconciled. Wall angles can now vary block by block across the pit. This is new, and we want to run it on real models. If your block model has slopes that change across the pit, bring it to first plan on us.

The sequence works the same way. Phases and bench precedence give the schedule coherent pools of blocks to work from rather than isolated picks spread across the pit. Within each month, diggers are assigned territories and walked through nearby blocks in a sensible working order. Haulage is solved in the same pass, so the fleet and the routes match the sequence instead of being fitted to it after the fact.

The plan moves equipment the way a crew would. That is a practical plan, and the difference between reviewing a schedule and rebuilding it.

Every plan comes out at monthly cadence, including life-of-mine. A ten-year schedule is 120 months of sequencing.

Raw block models run to millions of blocks, sometimes far more. No strategic scheduler works at that resolution, because no mine plan is executed at that resolution. Cognomine reblocks automatically before the solve, merging adjacent blocks onto a coarser grid.

What does not get cut is the constraint set: sequencing, dig precedence, grade targets, stockpile strategy, processing and haulage capacity, capital and operating assumptions. The trimming that usually happens to make a long-horizon model solvable is trimming of constraints, and that is the part we do not do.

The schedule comes back on the reblocked grid with coordinates, so it reimports into your tools. If you need to reconcile against the original model at its full resolution, that mapping happens on your side.

It also means the size of your raw model is not what sets solve time. Bringing your full model costs you nothing, and one credit still covers unlimited runs on it.

High Fidelity mode. Reblocking trades spatial resolution for solve time, and coarser scheduling units mean more waste travels with the ore. High Fidelity keeps the model closer to its original resolution, which produces a plan that mines less waste and is worth more. It takes longer, around an hour against under ten minutes for a standard run, so it suits the run you commit to rather than the scenarios you sweep through on the way there.

Two limits apply, and they measure different things. Cognomine accepts raw block models up to around one million blocks. Every model is then reduced to scheduling units before solving: under 100,000 in a standard run, up to 500,000 in High Fidelity. A model well inside the block limit is still reblocked, because the scheduling unit ceiling is what governs the solve.

High Fidelity is included in every credit. There is no additional charge and no separate entitlement.

If you already have a phased model, Cognomine works with it. If you do not, it recommends indicative phases and shows you which areas of the deposit hold the most value and should be prioritised.

That matters most on early-stage work, where phasing is one of the things the study is meant to determine. You do not need a finished phase design to get a schedule back.

Cognomine is built on a quantum-ready compute engine. It encodes a planning problem the way a quantum system would, then solves it on cloud GPUs. The speed is available now, on standard hardware, with a path to quantum processors as they mature and no change to how the problem is encoded. You do not need to know any of this to use Cognomine. It matters only because it is why the solve is fast.

Optimised phasing. More precise phase recommendations, driven by the economic sequencing of the pit rather than geometry alone.

This is in active development. It is not part of what you buy today, and we will tell you as soon as it lands.

Cognomine runs in the cloud, with nothing to install. Your data, and your clients' data, stays in Australia, is logically isolated to your account, and is removed within 30 days after your entitlement ends. The full data story, including encryption, retention, and standards, is on the security page.

Some enterprises need to run planning tools entirely inside their own environment, offline. If that is you, talk to us about what an on-site deployment would require.