PitForge — Exact Ultimate Pit Limit & Nested Whittle Pit-Shell Workbench
Published:
PitForge is an open-pit mine-design workbench. It solves the ultimate pit limit exactly, as a maximum-weight closure of the block-precedence graph reduced to a minimum cut on a Dinic max-flow engine, running live in the browser, and derives the nested Whittle pit shells by revenue factor. Live at pitforge.fasl-work.com, part of the Faena mining-analytics hub.
Say what the engine is
The exact result is the max-closure / min-cut equivalent of Lerchs-Grossmann, via Picard’s 1976 reduction on a Dinic engine. It is not a re-implementation of Lerchs-Grossmann, and it is not Hochbaum’s pseudoflow, which is referenced only as computing the same cut. From the exact pit, nested Whittle shells over an ascending revenue-factor schedule give value, tonnage and strip-ratio curves. Grade estimation runs three ways live (IDW, ordinary kriging, and a grade-nn ONNX surrogate).
Validated against MineLib, not against itself
The exact pit reproduces the published optima of three real MineLib instances: newman1 (1,060 blocks, 5.2 ms, relative error 9.96e-10), zuck_small (9,400 blocks, 237 ms, 1.86e-10), and kd (14,153 blocks / 219,778 precedences, 259 ms, 1.30e-10). All three match. Two further instances (marvin, mclaughlin_limit) are excluded with committed reasons rather than silently dropped. In real mode the scenario knobs are locked, because the instances publish their own net values and precedences and re-deriving them would break comparability with the published optimum.
The honest limit
Every solve asserts the duality identity pitValue = sum(positive) minus maxflow, so the optimiser checks itself. The learned grade-nn is a tie, not a win: R2 0.9613 versus ordinary kriging 0.958, a 0.0033 margin the artifact itself calls a fast approximation that never beats the exact result. Scheduling is a CPIT LP relaxation computed offline with scipy HiGHS, rendered from JSON, never live; it publishes a certified NPV upper bound with a 10.46 percent integrality gap, and states plainly that the rounded schedule is a heuristic and is never optimal. It is a design optimiser, not a JORC or NI 43-101 resource estimate.