Increase your Profits using

Advanced Process Control Technology

About 30 years ago scientists began researching the learning process that babies go through as they develop.  They learned that babies are born with a blank slate.  As the infant develops it collects sensory data which is used to make new brain connections and create a neural network (NN).  As the baby’s NN develops it continues to improve and refine itself.  Eventually the infant’s NN becomes so refined that the child can start doing things on their own with great precision without adult intervention.  Researchers then started trying to simulate the baby’s learning process with computers.  They figured out how to program a computer so that it can learn much the same way a baby does and create artificial neural networks (ANN).  Early in life a baby can only look at a soccer ball.  Within a few years the child develops skill until they can maneuver a soccer ball with skill.  Over the past 10 years ANN technology has proven to be a powerful tool for process optimization.  ANN has found very broad utility for optimizing processes from buying and selling stock to improving the efficiency of sophisticated chemical manufacturing processes. 

An area that is relatively untapped when it comes to the use of ANN is plastic fabrication and plastic manufacture.  There is a lot of expertise that goes into optimizing the performance of plastic fabrication and plastic manufacturing operations.  Whether you are making plastics, extruding plastic sheet, plastic films, and plastic foams or injection molding sophisticated plastic parts, there is a lot of tweaking that goes into optimizing the process; e.g., minimizing off-grade material, minimizing cycle time, minimizing the number of out of spec parts, maximizing energy  efficiency, maximizing strength properties, etc.  Our plastic experts can help you install ANN into your plastic fabrication and plastic manufacturing process to maximize efficiency, minimize off-grade, and maximize your profits.  The money it cost you to implement PlasticOptimizer™ technology into your processes will be regained within a few months of operation.

The following references showcase a few of the recent studies that have been done to demonstrate how ANN can help optimize plastic fabrication processes.

1) Modeling and optimization of a plastic thermoforming process.   Yang, Chyan; Hung, Shiu-Wan.    Journal of Reinforced Plastics and Composites  (2004),  23(1),  109-121.  Abstract: Thermoforming of plastic sheets has become an important process because of the relative low cost and good formability of plastic.  However, there are some unsolved problems that greatly effect that efficacy of the process.  Non-uniform thickness distribution caused by inappropriate processing conditions is one of them.  Experimental data were used to develop a neural network model for the thermoforming process via a supervised learning back propagation neural network. The model provided significant advantages in terms of improved product quality.

2) Optimization of injection molding process parameters.  S. Changyu, W. Lixia, L. Qian, C. Jingbo, ANTEC 2004 Proceedings, paper no. 112. Abstract: The process conditions in injection molding have critical influence on the part quality; so finding the optimum process parameters is the key to optimizing part quality. Application of artificial neural network technology is an effective tool for the process optimization during injection molding.

3) Advances in Injection Molding Process/Quality Control  Zhongbao Chen and Lih-Sheng Turng.   ANTEC 2004 Proceedings, paper no. 133.  Abstract:  Injection molding process/quality control is becoming more stringent due to increasing litigations involving failure of plastic parts. This paper reviews the state-of-the-art developments in injection molding quality control.  Real online quality control without human intervention has not yet been realized primarily due to lack of thorough understanding of the relationship among machine, process, and quality variables.  Neural network technology offers the capability of precision control during injection molding.

4) Artificial neural networks applied to polymer composites: a review.   Zhang, Z.; Friedrich, K.    Composites Science and Technology  (2003),  63(14),  2029-2044.  Abstract: Inspired by the human nervous system, an artificial neural network (ANN) approach is a fascinating math tool, which can be used to simulate a wide variety of complex scientific and engineering problems.  The objective of using ANNs is also to achieve the optimum design of composite materials for specific applications.  This paper reviews the principles of ANN development for optimizing plastic composite materials.  The properties that are optimized include fatigue life, wear performance, response under combined loading situations, and dynamic mechanical properties.  The goal of this review is to promote more consideration of using ANNs in the field of polymer composite property prediction and design.

5) Multi-objective optimization scheme for quality control in injection molding.   Liang, Jui-Ming; Wang, Pei-Jen.    Journal of Injection Molding Technology  (2002),  6(4),  331-342.  Abstract:  Optimization of process parameters for meeting the stringent quality requirements in injection molding has been studied via utilization of neural network modeling. 

6) Control of flow in resin transfer molding with real-time preform permeability estimation.   Nielsen, D. R.; Pitchumani, R.    Polymer Composites  (2002),  23(6),  1087-1110.  Abstract:  Variability in the preform structure in the mold are a challenge to achieving reliable preform molding processes.  Neural network modeling offers an effective means of deriving real-time process control decisions so as to steer the resin flow in a desired manner, which ensures excellent perform quality. 

The initial consultation is always free.  Email us today.

Email or call Plastic Failure Labs today toll free at 1-877-PLA-FAIL (1-877-752-3245)  or  (989) 385-2355