minehaulsim: A Deterministic Discrete-Event Simulator of Mine Haulage on a Constrained Road Network

The productivity of an open-pit mine is set at the truck-shovel interface, where a fleet of haul trucks cycles between shovels and dumps over a road network of ramps, junctions and single-lane segments. Simulating that interface faithfully needs the network: travel times depend on grade, and congestion (trucks bunching behind a slow unit on a one-way ramp) is where production is lost. Yet the open-source mine simulators use one fixed mine layout and scalar shovel-to-dump distances with no grades and no traffic, and the tools that model the network properly are commercial and closed. This software note presents minehaulsim, a pure-Python, MIT-licensed simulator that fills the gap: it runs mine haulage as a deterministic discrete-event simulation on a constrained road network (one-way ramps, width and passing classes, single-lane direction zones, junction blocking, and a headway-limited first-in-first-out no-overtake rule so that bunching emerges rather than being sampled from a distribution), with attainable truck speed solved from rimpull and retarder envelopes against grade. A run is a pure function of (spec, policy, seed) and is byte-identical across machines. Seeded parametric generators produce a structurally different, valid open-pit or multi-level underground mine per seed, and every run exports interoperable cycle logs. The simulator is demonstrated on a generated pit: production rises with the fleet and then saturates (from 243 to 4,800 tonnes per hour as the fleet grows to 22 trucks, the truck busy fraction falling from 1.0 to 0.70 as trucks begin to queue), the constrained network costs a 6-10% production loss over free-flow in the fleet-matched region, and the dispatch policy matters (shortest-queue 3,766 vs nearest-shovel 2,874 tonnes per hour at a matched fleet). The note is explicit about what the model is and is not; every number and figure regenerates from the committed run. Source and reproducible artifacts (MIT): https://github.com/fsantibanezleal/CAOS_MINEHAUL .