A hybrid intelligence technique based on the Taguchi method for multi-objective process parameter optimization of the 3D additive screen printing of athletic shoes

This paper presents a hybrid intelligence technique based on the Taguchi method for multi-objective process parameter optimization of 3D additive screen printing of athletic shoes. 3D additive screen printing is mainly used in the high-end athletic shoes and clothes field. It requires overlapping and overprinting dozens of times to make the printed patterns stereoscopic. The process of 3D additive screen printing is complex and variable and the production cycle is long. Because of the variability of the screen printing process and the coupling between process parameters, there is no simple method to guide the trial production of new products and obtain the optimal process parameters of screen printing. Trial-and-error is often used but it expends a lot of manpower, materials, and financial resources. To solve the optimization problem, a Taguchi experiment based on fuzzy comprehensive evaluation with five factors and two responses was first designed. Then, a back-propagation network (BPN), least-squares support-vector machine (LSSVM), and random forest (RF) were trained with experimental data to obtain a forecasting model for the process parameters. On comparison, the RF forecasting model performed best in this case. Then, the multi-objective antlion optimizer (MOALO), which is a new multi-objective optimization algorithm with excellent performance, was improved to the IMOALO, and it was proved that IMOALO has a better performance than MOALO. Combining the RF forecasting model with IMOALO, and carrying out the optimization, the optimal process parameters were obtained. Actual printing production shows that the proposed hybrid intelligence technique improves the production efficiency and first pass yield of printed products.

[1]  Marc P.Y. Desmulliez,et al.  The evolution of paste pressure during stencil printing , 2007 .

[2]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[3]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[4]  Seyed Mohammad Mirjalili,et al.  The Ant Lion Optimizer , 2015, Adv. Eng. Softw..

[5]  Madhan Shridhar Phadke,et al.  Quality Engineering Using Robust Design , 1989 .

[6]  H. T. Nagle,et al.  Utility of nonwovens in the production of integrated electrical circuits via printing conductive inks , 2008 .

[7]  Lyndon N. Smith,et al.  A neural network approach for selection of powder metallurgy materials and process parameters , 2000, Artif. Intell. Eng..

[8]  Z-C Lin,et al.  Combining the Taguchi Method with an Artificial Neural Network to Construct a Prediction Model for Near-Field Photolithography Experiments , 2010 .

[9]  Kok-Leong Ong,et al.  Prediction of Wool Knitwear Pilling Propensity using Support Vector Machines , 2010 .

[10]  R. Lippmann,et al.  An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[11]  Liang-Hua Chen,et al.  A bar-code recognition system using backpropagation neural networks , 1995 .

[12]  Pradeep Jangir,et al.  Multi-objective ant lion optimizer: a multi-objective optimization algorithm for solving engineering problems , 2016, Applied Intelligence.

[13]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[14]  Johan A. K. Suykens,et al.  Least Squares Support Vector Machine Classifiers , 1999, Neural Processing Letters.

[15]  S. Horvath,et al.  Unsupervised Learning With Random Forest Predictors , 2006 .

[16]  Bijay Ketan Panigrahi,et al.  Ant lion optimization for short-term wind integrated hydrothermal power generation scheduling , 2016 .

[17]  R. Durairaj,et al.  Critical factors affecting paste flow during the stencil printing of solder paste , 2001 .

[18]  E. S. Ali,et al.  Ant Lion Optimization Algorithm for Renewable Distributed Generations , 2016 .

[19]  BoulesteixAnne-Laure,et al.  Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics , 2012 .

[20]  Y. Zhong,et al.  Wool/cashmere identification based on projection curves , 2017 .

[21]  Yi-Shao Lai,et al.  Optimal design towards enhancement of board-level thermomechanical reliability of wafer-level chip-scale packages , 2007, Microelectron. Reliab..

[22]  Ndy Ekere,et al.  Sub process challenges in ultra fine pitch stencil printing of type‐6 and type‐7 Pb‐free solder pastes for flip chip assembly applications , 2005 .

[23]  Taho Yang,et al.  Design of manufacturing systems by a hybrid approach with neural network metamodelling and stochastic local search , 2002 .

[24]  Anne-Laure Boulesteix,et al.  Overview of random forest methodology and practical guidance with emphasis on computational biology and bioinformatics , 2012, WIREs Data Mining Knowl. Discov..

[25]  V. Vapnik.  The Support Vector Method of Function Estimation , 1998 .

[26]  Yize Sun,et al.  A model of screen reaction force for the 3D additive screen printing , 2018 .

[27]  Jaekyung Kim,et al.  Process parameter optimization of screen printing device for vacuum glazing pillar arrays , 2015 .

[28]  M.A. El-Sharkawi,et al.  Pareto Multi Objective Optimization , 2005, Proceedings of the 13th International Conference on, Intelligent Systems Application to Power Systems.