Monday, 31 October 2005: 12:30 PM-3:00 PM
302 (Cincinnati Convention Center)

Computers in Operations and Information Processing (10c)

#58 - Data Analysis: Design, Algorithms & Applications (10C06)
The avalanche of data made available by the continuing advances in experimental measurement techniques has created both opportunities and major challenges towards efficient computational processing of the data sets. Extracting useful information (i.e., knowledge) from these large data sets requires an integral approach that couples the design of experiments or measurement techniques with the algorithms or applications required for the data processing and analysis. New tools for representing, extracting, and using the knowledge available in data sets are expected to emerge to perform tasks in data-rich, analysis-rich situations rather than being data-poor, analysis-rich, which has been the traditional scenario until recently. Contributions are sought that propose new modeling approaches and computational algorithms for addressing issues related to data analysis, management and data-based decision making. Suggested topics include, but are not limited to: * Model based experimental design and use of new data to refine models * Analysis of complex systems including material and biological * Use of data analysis for design, optimization, and control * Interplay of first principles and data driven models in knowledge extraction. The following topics are also of interest to this session: * Distributed decision making in organizations * Decentralized information processing * Financial modeling & investment planning * Information management across the WWW. Applications with industrial relevance are strongly encouraged.
Chair:Santhoji Katare
12:30 PM58aData-Driven Soft Sensor Design - Application to Cement Kiln
Bao Lin, Bodil Recke, Jørgen Knudsen, Sten Bay Jørgensen
12:50 PM58bOvercoming False Positives (Type-I Errors) While Monitoring of Transient Operations Using Principal Component Analysis
Yew Seng Ng, Rajagopalan Srinivasan
1:10 PM58cAn Improved Methodology to Determine the Stochastic-Based Accuracy of Data Reconciliation-Based Estimators in Linear Systems
DuyQuang Nguyen, Miguel J. Bagajewicz
1:30 PM58dHierarchical K-Means Clustering Using Principal Components to Solve the Unsupervised Multi-Class Classification Problem
Syed B. Mohiddin, James Rathman, Chihae Yang
1:50 PM58eA Mixed-Integer Programming Approach to Multi-Class Data Classification Problem
Metin Turkay, Fadime Uney
2:10 PM58fValidation of a Model for a Biodiesel Production Process through Model-Based Experiment Design for Parameter Precision
Gaia Franceschini, Sandro Macchietto
2:30 PM58gHybrid Modeling and Multi-Objective Optimization of an Industrial Hydrocracker
Naveen Bhutani, Ajay K. Ray, Gade P. Rangaiah
Sponsor:Computing and Systems Technology Division

See more of Computing and Systems Technology Division

See more of The 2005 Annual Meeting (Cincinnati, OH)