|
Biscuit packet
filling
Fills (via Monte Carlo) cylindrical packets of biscuits. |
|
|
| wL,
wU |
g (specification limits) |
Packet weight lower and
upper specs. • |
| wμ,
wσ |
g |
Biscuit weight mean and st. dev.. |
| hU |
cm |
Packet height maximum. • |
| hμ,
hσ |
cm |
Biscuit height (thickness) mean and st. dev.. |
| Cost, penalties |
$/kg |
Cost of product ($/kg); weight and height penalties
(per bad packet). • |
| Items (N), .seed |
|
No. of biscuits, random no. gener. seed. • |
| tol, klass, ymax |
g−1
['0' (¬ '.0'), auto.] |
Tolerance, no. of histo. classes,
max. y for graph. • |
| Show values |
|
Shows the coordinates of the graph. • |
Simulates, via Monte Carlo,
the filling of cylindrical packets of biscuits.
The objective is to get packets with
weight within its (double) specification limits,
wL, wU, and
height below its (single) upper specification limit, hU.
The weight and height of each biscuit are considered
Gaussian. (Tolerance is for the inversion of the
Gaussian distribution.)
The computing time for 1 M (million) items is
~3 s (above 30 s it might be halted on the Internet).
The process parameter wU is possibly
the one to search for minimum loss, presented as loss per
packet processed.
|
| References: |
Plate: BiscuitPacket |
|
• Rhee, W., M.
Talagrand, 1991, A note on the selection of random variables under a sum constraint,
J. of Applied Probability, 28(4), 919–923 (vaguely related).
(Reserved ISI access. Abstract.pdf).
• Mettler Toledo, "Optimized Filling Quantities and Less Giveaway.
• 1909-07-15: Cochran, William Gemmell (1980-03-29). |
