Library / Systems Engineering Principles Framework
Jump to passage
In this reading

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 10:39:28 UTC · snapshot created 2026-10-03 10:40:04 UTC · last check 2026-10-03 11:15:11 UTC

5.1. Compare independent preparation and a transferable setting

A shorter stop can require a different physical arrangement, not just a different timetable. Consider one of the 25-minute changeovers. The machine works until minute 20. Preparation takes 12 minutes, installation and datum establishment 10, and the stipulated qualification operation 3. Performing all three after the stop gives the next start at minute 45.

An alternative prepares the next module on an independent station during minutes 8–20, then performs installation and qualification on the machine: the next start is minute 33. OPS.11.1 keeps the operator, station, module and machine demands in the same operating account. If the only fixture required for preparation remains occupied on the machine until minute 20, preparation cannot begin at minute 8 and this alternative returns to minute 45. A calendar change alone cannot create the missing independent means.

The SMED distinction between internal and external setup helps find that opportunity. A possible physical construction is a separately prepared module with a setting that can be transferred to the actual production pairing. The MIT kinematic-coupling proposal uses measured parameters of the mating halves and calibration against a known counterpart. It distinguishes the repeatability of one pair from interchangeability across pairs. Treat this as a design proposal whose actual geometry, contact, load, friction and deformation conditions must be established.

MMP.18 constructs the correspondence between the calibration and receiving descriptions; C.29.1 preserves the conditions under which the mathematical result can be used. In a translation-only illustration, a reference pairing places the module origin at 0.020 mm in the reference station’s coordinates. A module fiducial reads 10.020 mm there, so its coordinate relative to the module origin is 10.000 mm. The production pairing places the module origin at 0.035 mm in the production frame. The predicted fiducial coordinate there is 10.035 mm. Reusing the reference reading 10.020 mm would differ by 0.015 mm.

Suppose the module-coordinate calibration has a complete hard error bound of ±0.003 mm, already including uncertainty from the reference pairing, and the production-origin offset has a complete bound of ±0.002 mm. Their sum permits [10.030, 10.040] mm for the production coordinate. Whether this is adequate depends on the receiving operation. The calculation assumes the stated one-dimensional correspondence; it supplies neither the physical measurements nor a model of rotation, deformation or loss of contact.

With SYSE.26/.27, the proposed supported variant now states which prepared module and pairing it uses, which setting is returned and which conditions support that transfer. A different pairing or load that invalidates the correspondence returns the affected qualification question. The incumbent arrangement remains an alternative; a new coupling is unnecessary when a simpler arrangement already gives the required result.

The result is a candidate for reducing this stop from 25 to 13 minutes, with its resource condition, calibration correspondence and qualification needs exposed. The three-minute qualification operation is a stipulated duration, not evidence that three minutes can qualify an actual process. SYSE.24 compares the complete provision alternatives before adopting the change. The rest of the day’s capacity and coating/return constraints still apply.