The Crew and the Machine

Abstract

Every capital model for an automation deployment rests on an assumption about the duty cycle: the machine will run the hours it was designed to run. That assumption is not a property of the machine — it is a property of the crew around it, and in the model developed here an unmanaged workforce response converts a twenty-one-month payback into a fifty-six-month one without a single act of overt resistance. The paper's central distinction is between two fears that wear the same uniform on a shop floor: an uncalibrated threat response, which communication genuinely cures, and a calibrated reading of a real personal distribution — the firm holds a portfolio and can absorb variance; the worker holds a single position and cannot — which communication does not reduce but merely converts from open resistance to covert, strictly worse for the asset because a supervisor cannot manage what he cannot see. The analysis audits what the loss-aversion literature actually supports, sets out a six-mechanism taxonomy of utilization leakage with observable signatures, reads the German robot-adoption record as a natural experiment in incumbent protection, prices the liability in a sixty-month model, and specifies the commitment architecture that changes a worker's actual distribution rather than his perception of it — real money, spent before there is evidence it was necessary, belonging in the capital request rather than a change-management line nobody reads. The fourth paper in the robotics deployment series: REA·004 gave the asset thesis, REA·005 the deployment floor, REA·006 the capital structure — this one prices the human factor the other three quietly assumed.