|
Activity Number:
|
556
- Essentials of Statistics for Advanced Manufacturing Quality
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, August 2, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Quality and Productivity Section
|
|
Abstract #322295
|
|
|
Title:
|
A Mixed Variance Component Model for Quantifying the Elasticity Modulus of Nanomaterials
|
|
Author(s):
|
Xinwei Deng* and Angang Zhang
|
|
Companies:
|
Virgiania Tech and Virgiania Tech
|
|
Keywords:
|
mixed variance ;
nano-quantification ;
group variable selection
|
|
Abstract:
|
Nanomaterials possess great mechanical properties with wide applications in many areas. Experiments are often conducted for measuring certain mechanical properties of interest. How to accurate quantify mechanical properties of nanomaterials is thus very important but challenge due to nanoscale manipulation and tactful measurement techniques. Statistical modeling approach combined physical theories have been used for the quantification of nanomaterials. In this work, we propose a novel mixed variance component model to accommodate experimental variations and artifacts for analyzing the nanomaterial experiment data. The proposed method can automatically adjust systematic errors occurred in the experiments through a group adaptive forward backward selection (GFoBa). It thus leads to accurate estimation of mechanical properties with the ability to filter out various experimental errors. The performance of the proposed method is compared with other existing method through both simulation and a real data example.
|
Authors who are presenting talks have a * after their name.
