Method for Product Design Time Forecasting Based on Support Vector Regression with Probabilistic Constraints

There exist problems of small samples and heteroscedastic noise in design time forecast. To solve them, support vector regression with probabilistic constraints (PC-SVR) is proposed in this article. The mean and variance functions are simultaneously constructed based on a heteroscedastic regression model. Probabilistic constraints are designed to make sure that for every sample, the forecast value is in a neighborhood of the target value with high probability. The optimization objective is formatted in the form of par-v-SVR. Prior knowledge about maximum completion time can be embedded in probabilistic constraints, and provides the size of the neighborhood of the target value. The results of application in injection mold design have confirmed the feasibility and validity of PC-SVR.

[1]  Jude W. Shavlik,et al.  Knowledge-Based Kernel Approximation , 2004, J. Mach. Learn. Res..

[2]  Doheon Lee,et al.  Possibilistic support vector machines , 2005, Pattern Recognit..

[3]  Zheng Wang,et al.  Computing Completion Time and Optimal Scheduling of Design Activities in Concurrent Product Development Process , 2010, IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans.

[4]  Steven D. Eppinger,et al.  A simulation-based process model for managing complex design projects , 2005, IEEE Transactions on Engineering Management.

[5]  Hamdi A. Bashir,et al.  Models for estimating design effort and time , 2001 .

[6]  Margarida F. Jacome,et al.  NREC: Risk Assessment and Planning of Complex Designs , 1997, IEEE Des. Test Comput..

[7]  Bernhard Schölkopf,et al.  New Support Vector Algorithms , 2000, Neural Computation.

[8]  Tao Yao,et al.  An overlapping-based process model for managing schedule and cost risk in product development , 2012, Concurr. Eng. Res. Appl..

[9]  Jacques Wainer,et al.  A COMBINATION OF SUPPORT VECTOR MACHINE AND k-NEAREST NEIGHBORS FOR MACHINE FAULT DETECTION , 2013, Appl. Artif. Intell..

[10]  Olvi L. Mangasarian,et al.  Nonlinear Knowledge in Kernel Approximation , 2007, IEEE Transactions on Neural Networks.

[11]  Howard Huang,et al.  Process knowledge based multi-class support vector classification (PK-MSVM) approach for surface defects in hot rolling , 2011, Expert Syst. Appl..

[12]  Shun-Feng Su,et al.  Robust support vector regression networks for function approximation with outliers , 2002, IEEE Trans. Neural Networks.

[13]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[14]  Jonathan Cagan,et al.  The role of timing and analogical similarity in the stimulation of idea generation in design , 2008 .

[15]  Han-Xiong Li,et al.  A probabilistic support vector machine for uncertain data , 2009, 2009 IEEE International Conference on Computational Intelligence for Measurement Systems and Applications.

[16]  Abbie Griffin,et al.  Modeling and measuring product development cycle time across industries , 1997 .

[17]  Duo Xu,et al.  An Approach to Estimating Product Design Time Based on Fuzzy $\nu$-Support Vector Machine , 2007, IEEE Transactions on Neural Networks.

[18]  Washington Okori,et al.  CONTRIBUTION OF PRIOR KNOWLEDGE TO PROBABILISTIC PREDICTION OF FAMINE , 2013, Appl. Artif. Intell..

[19]  D. Xu,et al.  An intelligent estimation method for product design time , 2006 .

[20]  Pei-Yi Hao,et al.  New support vector algorithms with parametric insensitive/margin model , 2010, Neural Networks.