A 0。0026 1 0。0026 127。80 <0。0001

D 0。0010 1 0。0010 51。08 <0。0001

AD 0。0002 1 0。0002 10。79 0。0054

Residual 0。0003 14 0。00002

Total 0。0042 17

Table 4d

ANOVA results of  dyexp。

lected as a desirability function form for minimizing the warpage distortion and the mass loss。 Furthermore, these inpidual desir- ability functions are can be calculated from Eq。 (1) as a combined desirability function。 More details can be  found  in  the  Myers’s and Montgomery book (Myers & Montgomery,  2002)。

5。1。 ANOVA results

The analysis of variance (ANOVA) is conducted and the results of simulation and experimental are shown  in  Tables  4a–4c and 4d, respectively。 A ‘‘Model F value” is calculated from a model mean square pided by a residual mean square。 It is a test of com- paring a model variance with a residual variance。 If the variances are close to the same, the ratio will be close to one and it is less likely that any of the factors have a significant effect on the re- sponse。 As for a ‘‘Model P value”, if the ‘‘Model P value” is very small (less than 0。05) then the terms in the model have a signifi- cant effect on the response (Myers & Montgomery, 2002)。 Simi- larly, an ‘‘F value” on any inpidual factor terms  is  calculated from a term mean square pided by a residual mean square。 It is a test that compares a term variance  with a  residual variance。 If the variances are close to the same, the ratio will be close to one and it is less likely that the term has a significant effect on the re- sponse。 Furthermore, if a ‘‘P value” of any model terms is very small (less than 0。05), the inpidual terms in the model have a sig- nificant effect on the  response。

5。1。1。 ANOVA results for simulation

Table 4a lists the ANOVA result of the dxsim。 A ‘‘Model F value” of 3246。14 with a ‘‘Model P value” of less than 0。0001 suggested that the selected model is significant。 A ‘‘P value” for the model term ‘‘A” (the melt temperature) and ‘‘D” (the packing pressure) are less than 0。0001, which are less than 0。05, indicating that the model term ‘‘A” and ‘‘D” are significant。 Additionally, two interac- tion terms, ‘‘AC”, and ‘‘CD”, also have significant influences to the dxsim, namely, through ANOVA, the significant terms that have sig- nificant impacts on the injection molding process can be identified。 Table 4b shows the ANOVA result of the dysim。 A ‘‘Model F value” of

171。94 with a ‘‘Model P value” of less than 0。0001 implies that   the

selected model is significant。 A ‘‘P value” for the model term ‘‘A” (the melt temperature) is less than 0。0001, which  is  less  than 0。05, indicating that the model term ‘‘A” is significant。 Similarly, the model term ‘‘C” and ‘‘D” (the inject speed and the packing pres- sure, respectively) are significant。 Additionally, a interaction terms ‘‘AD”, also have significant influences to the    dysim。