Data-driven optimization of dynamic reconfigurable systems of systems.

This report documents the results of a Strategic Partnership (aka University Collaboration) LDRD program between Sandia National Laboratories and the University of Illinois at Urbana-Champagne. The project is titled 'Data-Driven Optimization of Dynamic Reconfigurable Systems of Systems' and was conducted during FY 2009 and FY 2010. The purpose of this study was to determine and implement ways to incorporate real-time data mining and information discovery into existing Systems of Systems (SoS) modeling capabilities. Current SoS modeling is typically conducted in an iterative manner in which replications are carried out in order to quantify variation in the simulation results. The expense of many replications for large simulations, especially when considering the need for optimization, sensitivity analysis, and uncertainty quantification, can be prohibitive. In addition, extracting useful information from the resulting large datasets is a challenging task. This work demonstrates methods of identifying trends and other forms of information in datasets that can be used on a wide range of applications such as quantifying the strength of various inputs on outputs, identifying the sources of variation in the simulation, and potentially steering an optimization process for improved efficiency.

[1]  Panos Y. Papalambros,et al.  Analytical Target Cascading in Automotive Vehicle Design , 2001 .

[2]  Jiawei Han,et al.  Classifying large data sets using SVMs with hierarchical clusters , 2003, KDD '03.

[3]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[4]  Conrad S. Tucker,et al.  Data-Driven Decision Tree Classification for Product Portfolio Design Optimization , 2009, J. Comput. Inf. Sci. Eng..

[5]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[6]  Conrad S. Tucker,et al.  Optimal Product Portfolio Formulation by Merging Predictive Data Mining With Multilevel Optimization , 2008 .

[7]  Mario Drobics,et al.  Creating comprehensible regression models , 2007, Soft Comput..

[8]  Yuanhui Zhang,et al.  A ReliefF attribute weighting and X-means clustering methodology for top-down product family optimization , 2010, Engineering Optimization.

[9]  Aiko M. Hormann,et al.  Programs for Machine Learning. Part I , 1962, Inf. Control..

[10]  Roger Jianxin Jiao,et al.  Product portfolio identification based on association rule mining , 2005, Comput. Aided Des..

[11]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[12]  Ramakrishnan Srikant,et al.  Fast Algorithms for Mining Association Rules in Large Databases , 1994, VLDB.

[13]  Thaddeus Tarpey,et al.  A parametric k-means algorithm , 2007, Comput. Stat..

[14]  Ian H. Witten,et al.  The WEKA data mining software: an update , 2009, SKDD.

[15]  John A. Hartigan,et al.  Clustering Algorithms , 1975 .