A Critical Review of the Logistics and Construction Methods for the Great Pyramid Using a Discrete-Event Digital Twin

Almost every proposal about the Great Pyramid is a proposal about the lift. Straight ramps, spiral ramps, ramps folded onto the faces, levers, rockers, counterweights sliding down the Grand Gallery: the argument is always about how a block travelled the last hundred and forty metres. This paper asks a different question.

If the whole chain from the quarry face to the setting crew is simulated as coupled state machines, with each stage given a fixed crew and the calendar left to emerge, which stage actually sets the pace? The instrument is a twin of the building process, not a survey model of the building. The answer is the quarry, and it is not close.

In every configuration that finishes inside Khufu’s reign, the lift has spare capacity on roughly nine working days in ten: it could have raised more stone than the quarry delivered. Three consequences follow. First, the admissible builds are not a scatter of estimates but a single curve: quarrymen multiplied by extraction rate must equal 283 cubic metres of finished block a working day, and the Giza quarry’s own frontage truncates that curve at 0.047 cubic metres per man-day.

Below that rate there is no face for the men to stand on. The one experimental rate in the literature, about 0.05 cubic metres of block per man-day from copper-tool trials at Wadi el-Jarf, sits on that floor. The floor rests on an estimate of the working frontage, 4,000 metres at 1.5 men to the metre, which is a construction of ours and the most fragile number here; the qualitative claim survives across a wide range of it, the precise figure does not.

Second, because the lift has slack, the schedule cannot choose between lifting schemes. A multi-ramp, tongues of laid block, and a straight ramp stopped at twenty-five metres all reach the capstone within three months of one another, and a steeper ramp saves earthwork without saving a single day. The one exception is the speed of a lever station: above roughly thirty minutes a block a course, no number of lever crews finishes the pyramid, which rules out cribbing at Isler’s measured rate while leaving the rocker and the A-frame comfortably inside the reign.

There is a deeper reason the calendar is powerless here. The gravitational work of assembling the pyramid is about 2.1 TJ, which is under two per cent of the labour the site actually fields, and about seven per cent once sledge friction is included. No lifting machine can save more than that, and a counterweight cannot save even that, because a pyramid is a one-way flow of mass upward and there is nothing to send down to reset it.

Third, what does discriminate is not the calendar but the spoil. The schemes that agree on the schedule differ by five orders of magnitude in ramp fill, from nothing at all to 5.38 million cubic metres, and Lehner has already proposed the test: the debris now filling the Khufu quarry may be the dismantled ramps. Nobody has measured its volume.

We set out what that measurement would settle. Simulation invites thinking and validation. It is not a proof in itself.