Data-driven model for estimation of friction coefficient via informatics methods

Eric W. Bucholz; Chang Sun Kong; Kellon R. Marchman; W. Gregory Sawyer; Simon R. Phillpot; Susan B. Sinnott; Krishna Rajan

doi:10.1007/s11249-012-9975-y

Data-driven model for estimation of friction coefficient via informatics methods

Eric W. Bucholz, Chang Sun Kong, Kellon R. Marchman, W. Gregory Sawyer, Simon R. Phillpot, Susan B. Sinnott, Krishna Rajan

Materials Science and Engineering

Research output: Contribution to journal › Article › peer-review

20 Scopus citations

Abstract

As technologies progress, the development of new mechanical systems demands the rapid determination of friction coefficients of materials. Data mining and materials informatics methods are used here to generate a predictive model that enables efficient high-throughput screening of ceramic materials, some of which are candidate high-temperature, solid-state lubricants. Through the combination of principal component analysis and recursive partitioning using a small dataset comprised of intrinsic material properties, we develop a decision tree-based model comprised of if-then rules which estimates the friction coefficients of a wide range of materials. This datadriven model has a high degree of accuracy with an R ² value of 0.8904 and provides a range of possible friction coefficients that accounts for the possible variability of a material's actual friction coefficient.

Original language	English (US)
Pages (from-to)	211-221
Number of pages	11
Journal	Tribology Letters
Volume	47
Issue number	2
DOIs	https://doi.org/10.1007/s11249-012-9975-y
State	Published - Aug 2012

All Science Journal Classification (ASJC) codes

Mechanics of Materials
Mechanical Engineering
Surfaces and Interfaces
Surfaces, Coatings and Films

Access to Document

10.1007/s11249-012-9975-y

Fingerprint Dive into the research topics of 'Data-driven model for estimation of friction coefficient via informatics methods'. Together they form a unique fingerprint.

Cite this

@article{a772cd81786e4f248935ff9b4268bfed,

title = "Data-driven model for estimation of friction coefficient via informatics methods",

abstract = "As technologies progress, the development of new mechanical systems demands the rapid determination of friction coefficients of materials. Data mining and materials informatics methods are used here to generate a predictive model that enables efficient high-throughput screening of ceramic materials, some of which are candidate high-temperature, solid-state lubricants. Through the combination of principal component analysis and recursive partitioning using a small dataset comprised of intrinsic material properties, we develop a decision tree-based model comprised of if-then rules which estimates the friction coefficients of a wide range of materials. This datadriven model has a high degree of accuracy with an R 2 value of 0.8904 and provides a range of possible friction coefficients that accounts for the possible variability of a material's actual friction coefficient.",

author = "Bucholz, {Eric W.} and Kong, {Chang Sun} and Marchman, {Kellon R.} and Sawyer, {W. Gregory} and Phillpot, {Simon R.} and Sinnott, {Susan B.} and Krishna Rajan",

note = "Funding Information: Acknowledgments The authors EWB, KRM, WGS, SRP, and SBS gratefully acknowledge the support of the Office of Naval Research under grant number N000141010165. CSK and KR acknowledge the support from the NSF-ARI Program under grant number CMMI 09-389018 and the Army Research Office under grant number W911NF-10-0397. KR acknowledges the support from the Wilkinson Professorship of Interdisciplinary Engineering. The authors also thank Jonathan Liddy, former undergraduate student from the University of Florida, for his role in the compilation of the properties from the literature used for the material dataset in this study.",

year = "2012",

month = aug,

doi = "10.1007/s11249-012-9975-y",

language = "English (US)",

volume = "47",

pages = "211--221",

journal = "Tribology Letters",

issn = "1023-8883",

publisher = "Springer New York",

number = "2",

}

TY - JOUR

T1 - Data-driven model for estimation of friction coefficient via informatics methods

AU - Bucholz, Eric W.

AU - Kong, Chang Sun

AU - Marchman, Kellon R.

AU - Sawyer, W. Gregory

AU - Phillpot, Simon R.

AU - Sinnott, Susan B.

AU - Rajan, Krishna

N1 - Funding Information: Acknowledgments The authors EWB, KRM, WGS, SRP, and SBS gratefully acknowledge the support of the Office of Naval Research under grant number N000141010165. CSK and KR acknowledge the support from the NSF-ARI Program under grant number CMMI 09-389018 and the Army Research Office under grant number W911NF-10-0397. KR acknowledges the support from the Wilkinson Professorship of Interdisciplinary Engineering. The authors also thank Jonathan Liddy, former undergraduate student from the University of Florida, for his role in the compilation of the properties from the literature used for the material dataset in this study.

PY - 2012/8

Y1 - 2012/8

N2 - As technologies progress, the development of new mechanical systems demands the rapid determination of friction coefficients of materials. Data mining and materials informatics methods are used here to generate a predictive model that enables efficient high-throughput screening of ceramic materials, some of which are candidate high-temperature, solid-state lubricants. Through the combination of principal component analysis and recursive partitioning using a small dataset comprised of intrinsic material properties, we develop a decision tree-based model comprised of if-then rules which estimates the friction coefficients of a wide range of materials. This datadriven model has a high degree of accuracy with an R 2 value of 0.8904 and provides a range of possible friction coefficients that accounts for the possible variability of a material's actual friction coefficient.

AB - As technologies progress, the development of new mechanical systems demands the rapid determination of friction coefficients of materials. Data mining and materials informatics methods are used here to generate a predictive model that enables efficient high-throughput screening of ceramic materials, some of which are candidate high-temperature, solid-state lubricants. Through the combination of principal component analysis and recursive partitioning using a small dataset comprised of intrinsic material properties, we develop a decision tree-based model comprised of if-then rules which estimates the friction coefficients of a wide range of materials. This datadriven model has a high degree of accuracy with an R 2 value of 0.8904 and provides a range of possible friction coefficients that accounts for the possible variability of a material's actual friction coefficient.

UR - http://www.scopus.com/inward/record.url?scp=84864671596&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864671596&partnerID=8YFLogxK

U2 - 10.1007/s11249-012-9975-y

DO - 10.1007/s11249-012-9975-y

M3 - Article

AN - SCOPUS:84864671596

VL - 47

SP - 211

EP - 221

JO - Tribology Letters

JF - Tribology Letters

SN - 1023-8883

IS - 2

ER -

Data-driven model for estimation of friction coefficient via informatics methods

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Cite this