One order is not one print
A twelve-part order printed three parts at a time needs four successful plates. Six parts per plate need two. The larger plate may use less support material or less machine time overall, but that is a slicer-dependent input, not an automatic economy of scale. Run each layout through your slicer first, then compare actual plate totals here.
The three quantities that matter
Delivered parts are what the customer needs. Parts per plate describe a successful full print. Successful plates are the delivered quantity divided by plate capacity, rounded up. At five requested parts and three parts per plate, the model costs two plates and flags one spare place.
Successful plates = ceiling(delivered parts / parts per plate)
Worked example: six parts, two layouts
In an illustrative material-only example, a two-part plate uses 100 g and a three-part plate uses 140 g. With a 1,000 g spool costing 20 currency units, layout A uses three plates and costs 6.00. Layout B uses two plates and costs 5.60. That is only a 0.40 material saving; a slower layout or longer finishing can reverse the overall result.
Allocate labour once, not twice
Order setup is charged once. Finishing is charged per delivered part. Retry handling is charged per expected failed attempt. If you have routine handling for every successful plate, include its total in setup for that scenario. The model does not add it automatically. When comparing different batch quantities, review setup again rather than assuming it remains fixed.
The partial-plate limitation
This version deliberately charges a full final plate and does not credit surplus output. For a partially filled final plate with a different slicer estimate, calculate it separately and avoid duplicating order setup. A future mixed-plate planner should model this explicitly, rather than silently prorating time by part count.