00337 �cc2c711f7435285954e5a4f8b72a11

340

McCarville & Montgomery

When applying the optimum guard bands to the defective level model and the Gage Loss model, Equations (9) and (14), the defective level becomes 105.663 ppm and the gage loss becomes 74371.7 ppm, or 7.437%.

Numerical Optimization

Naturally, with morę than two gages, the graphical optimization method becomes morę difficult. Numerical methods can be utilized to optimize these linear models. Nonlinear programming techniąues are an obvious choice where the constraint model is the ppm(Def) and the model to be minimized is the Gage Loss. G. C. Derringer and R. Suich (1980) developed a nonlinear programming method that is ąuite effective for solving problems such as these. They suggest that by developing a single measure of desirability and by finding the maximum desirability level, an optimal solution to the problem will be found.

For example, if the objective for a response variable is to achieve a target of Bj while staying within the specification limits of Ai and Ci such that Ai²Bi²Q, then the desirability would increase from Ai to Bj and decrease from Bi to Q. The desirability functions for optimizing around a target as well as minimizing and maximizing are shown in Figurę 19 along with the associated effects of their weights.

Once the desirability, dj, is determined for each of m responses, an overall desirability is calculated as follows:

D = (d₁d₂d₃...d_m)¹/^m

The overall desirability can now be optimized by finding its maximum value. Design-Expert utilizes a simplex steepest ascent method for finding the optimum. A starting point is randomly selected and the steepest ascent begins. Several optimizations should be performed starting from randomly selected points in the design region. Design-Expert performs this automatically. If the Solutions are fairly close together, then one optimum exists within the region; however, if the Solutions are widely different, then multiple optimals exist.

With all of the weights set at one, ten cycles of non-linear programming was performed which yielded optimal values of 642.5 for GB1 and 613.4 for GB2. The predicted values for the ppm(Def) and Gage Loss are 100 and 74,600 respectively. When applying these values for GB1 and GB2 to the original models developed earlier, Eąuations (9) and (14), the calculated ppm(Def) becomes 99.9988 and the Gage Loss is 75,162.6, or 7.516%.

Wyszukiwarka

Podobne podstrony:
dctrimcntal effea ©o pubłic opiiuon at a rirac when 1t was im poeta ni to emure a iceling or teamy a
00335 g3b40dbbf66d5e5072621ad7f05eef6 33 8 McCarville & Montgomery Graphical Optimization When
00196 ?c3e9ff6e1645b5c66f96f0e3b703fa 198 Messina, Montgomery, Keats & Runger The residual for
00313 N8ca2ce47ae6338a7cbe578cf96f640 316 McCaryille & Montgomery Figurę 1. Probabiiity of xj,,
00323 ?85e6f46f7f32f30fdf82adebd3e7fe 326 McCarviIle & Montgomery Figurę 10. The Defect Level D
00329 ?f4634893f39e8884db8751e17aa760 332 McCarville & Montgomery 0.136772. This becomes a pote
00339 +06a963b1e1eb1e683a249968625de1 342 McCarville & Montgomery Additionał Considerations The
"^SAFETY PRECAUTIONS* (Always read these instructions before the exercise.) When designing the
PPTP When defining the PPTP server s IP rangę, avoid overlap with the rangę of IP addresses handed o
IMGR33 (2) GILDING AND TECHNIQUES OF DECORATION sunulatcd in che woodcn moulds. When complctcd the w
Kavicsfestés I 37 When 1 tofd my friends I was going to give them cooties, you shouJd have seen th
THE STRUCTURE OF CLUSTERSFROM POTATO AMYLOPECTIN 113 When considering the extensive number of irwest
IB1 Hołd the areas closest to the shoulder. *ands clench to fists when :.i : a detaining The c
CLILPYRAMID of Quality CUL based on the 4Cs-Framew,ork can only be achieved when all the four C: are

więcej podobnych podstron