a b s t r a c t   This paper deals with the application computer-aided engineering integrating with statistical technique to reduce warpage variation depended on injection molding process parameters during production of thin-shell plastic components。 For this purpose, a number of Mold-Flow analyses are carried out by uti- lizing the combination of process parameters based on three level of L18 orthogonal array table。 In the meantime, apply the design of experiments (DOE) approach to determine an optimal parameter setting。 In addition, a side-by-side comparison of two different approaches of simulation and experimental is pro- vided。 In this study, regression models that link the controlled parameters and the targeted outputs are developed, and the identified models can be utilized to predict the warpage at various injection molding conditions。 The melt temperature and the packing pressure are found to be the most significant factors in both the simulation and the experimental for an injection molding process of thin-shell plastic   parts。 72685

1。 Introduction

The injection molding is an important manufacturing process to polymers; it provides products with high dimensional steadiness, low manufacture cycles and low costs。 Therefore, injection mold- ing is a widely used process for polymeric  materials  (Brent, 2000)。 Many studies all found that injection molding processing parameters have crucial effects on the quality of products (Chen, Tai, Wang, Deng, & Chen, 2008; Chien, Chen, Lee, & Huang, 2004; Ismail & Suryadiansyah, 2004; Lin, Deng, Huang, & Yang, 2008; Ok- tem, Erzurumlu, & Uzman, 2007; SadAbadi & Ghasemi, 2007)。 However, various industries have employed the Taguchi method over the years to improve products or manufacturing processes。 It is a powerful and effective method to solve challenging quality problems。 Actually, the design of experiments (DOE) method has been used quite successfully in several industrial applications like in optimizing manufacturing processes or designing electrical/ mechanical components (Puertas & Luis, 2004; Sofuoglu, 2006; Tong, Tsung, & Yen, 2004; Yang, 2006a; Yang,  2006b)。

Recently, computer-aided engineering (CAE) has been success- fully used in the simulation of the injection molding process, since it provides designers/engineers with visual and numerical feed- back of the part behavior and eliminates the traditional trial and error approach for optimization。 Proper interpretation of the re- sults from simulation can help selecting a suitable material; reduce

cycle time and costs on mold alteration。 Lotti, Ueki, and Bretas (2002) used Mold-Flow simulation to investigate the parameters influencing shrinkage of polypropylene plaques。 They found that the holding pressure and the mold temperature contributed signif- icantly to the shrinkage of the parts。 Chen, Cheng, Wang, and Chien (2005) had used a software, Mold-Flow and Taguchi method to simulate the process of gas assisted injection molding to find out the optimal design condition of manufacturing a polystyrene prod- uct。 It is found that a slower gas injection speed, higher melt tem- perature, higher gas pressure, and longer gas packing time    would

yield lesser warpage。 Patcharaphun, Zhang, and Mennig (2007) performed the aid of a commercial software package (Moldflow®) to predict the fiber orientation distribution within the weldline area of push–pull-processing parts。 The predicted values of orien- tation tensor components are found to agree reasonably well with corresponding experimental measurements。 Patcharaphun and Mennig (2007)  a  commercial  simulation  package (Moldflow®) has been extensively compared with experiments in sandwich injection molding process。 A good agreement between simulation and experimental results indicates that the Moldflow®  program can be used as a valuable tool for the prediction of melt-flow behavior during the sandwich injection process。 Song, Liu, Wang, Yu, and Zhao (2007) had applied the orthogonal experiment meth- od (Taguchi method) and numerical simulation software Mold- flow®, discussed the influence of different process parameters (injection rate, injection pressure, melt temperature, metering size and part thickness) on the molding process for ultra-thin wall plas- tic parts, and so  on。