Abstract Energy efficiency is an essential consideration in sustainable manufacturing。 This study presents the car fender-based injection mold- ing process optimization that aims to resolve the trade-off between energy consumption and product quality at the same time in which process parameters are optimized variables。 The process is specially optimized by applying response surface methodology and using non- dominated sorting genetic algorithm II (NSGA II) in order to resolve multi-object optimization problems。 To reduce computational cost and time in the problem-solving procedure, the combination of CAE-integration tools is employed。 Based on the Pareto diagram, an ap- propriate solution is derived out to obtain optimal parameters。 The optimization results show that the proposed approach can help effec- tively engineers in identifying optimal process parameters and achieving competitive advantages of energy consumption and product quality。 In addition, the engineering analysis that can be employed to conduct holistic optimization of the injection molding process in order to increase energy efficiency and product quality was also mentioned in this paper。75746
Keywords: Multi-objective optimization; Injection molding process; Energy efficiency; Plastic car fender
1。 Introduction
Injection molding has been the most popular method for making plastic products due to high efficiency and manufac- turability。 The injection molding process includes four impor- tant stages: filling, packing, cooling, and ejection。 A primary traditionally concern in injection molding has been that of the extent to which high-quality products with strong mechanical properties can be manufactured, in the absence of any unde- sired defects。 Many previous studies have sought to eliminate defects in plastic products。 To minimize temperature devia- tion for an automotive product, cooling circuit parameters were optimized using CAE program [1]。 In the case of reduc- ing warpage, the combination of the response-surface method or neural network with a genetic algorithm was conducted in order to obtain optimal parameters [2, 3]。 The deficiency of these studies was the lack of consideration for the energy consumption of the injection molding process。 Consumer pressure, rising energy cost and environmental legislation have combined to increase the importance of reducing energy consumption in the industrial plastic industry。
To enhance practical application, both energy consumption
*Corresponding author。 Tel。: +82-52-259-2294, Fax。: +82-52-259-1680 E-mail address: phosk@ulsan。ac。kr
© Society of CAD/CAM Engineers & Techno-Press doi:10。7315/JCDE。2014。025
and product quality should be taken into consideration。 Ener- gy saving for the injection molding process can be pided into two sub-aims。 In the first sub-aim, companies and manu- facturers focus on machine improvement and manufacturing technologies related to injection molding machine hardware and auxiliary equipment。 The second sub-aim focuses on optimization of process parameters in the operating process, to reduce energy consumption。 Whereas in the first sub-aim, adoption of new-generation or rebuilt machines with ad- vanced energy-saving devices is very expensive, a much lower cost is required in the optimization-based second aim which only requires experimental or simulated data。 In this sub-aim, a mathematical model among process parameters and an energy model are established, based on supplied data。
Energy-saving via process parameter optimization has at- tracted much research attention。 By using analytical method or artificial neural network (ANN), the interrelationship be- tween process parameters and energy consumption was es- tablished [4, 5]。 However, among these studies, product qual- ity was lacking。
To reduce energy-consumption and carbon emissions, while increasing product quality in the injection molding process of plastic car fender, this paper proposes a multi- objective optimization framework that addresses multiple considerations in the process。 The paper focuses on optimiz-