Abstract
This study proposes a procedure that integrates the grey relational analysis and principal component analysis (PCA) to optimize a multi-response problem. The procedure is demonstrated by investigating the energy at failure in single impact, and the impact fatigue life of the injection molded 30 wt% glass fiber reinforced poly(butylene terephthalate) (PBT). The results show that the influence of the injection molding condition on the product’s impact performances are quite significant. The signal-to-noise (S/N) ratios of impact performances of the recommended injection molding condition are larger than the averages of experimental runs. This demonstrates that the proposed procedure is effective for the optimization of multi-response problems.
