Hybrid machine learning framework for the classification and prediction of dry sliding wear performance of friction stir processed and laser shock peened AA6061-O alloy

This study presents a hybrid machine learning model (classification and regression) based framework to understand the tribological properties of Aluminium alloy 6061 (AA6061). At first hybrid experimental approach combining friction stir processing (FSP) and laser shock peening (LSP) was conducted with the AA6061-O alloy (base material). For this, wear rate (W R ) of AA6061 is observed as a function of variable applied load (AL - 10, 20, and 30 N), variable sliding velocity (SV - 1, 2, and 3 m/s) and constant sliding distance (SD – 1000 m). Analysis of variance (ANOVA) showed that AL and SV significantly influenced W R (p < 0.05), contributing more than 89% to the total variation. Results obtained from the real-time experiments are then used to train, test and validate both machine learning (ML) models. Results recorded by the J48 decision tree proven its balanced classifying ability across both HIGH and LOW classes. The weighted average F-measure of 0.892 for BM, 0.886 for FSP, and 0.886 for LSP. Meanwhile, results recorded by the extreme gradient boosting (XGBoost) validated its combined gradient boosting nature with regularization and effective handling of nonlinear relationships, and strong generalization. Also, this study attempts the generalization of the proposed model through the model interpretability study. Results of interpretability analysis confirms SV as the first and AL as the second order influential parameters in governing the wear rate. Thus, the proposed hybrid framework offers a reliable and effective tool for understanding the tribological behavior of AA6061.

Authors

Institutions

Publication Details

Journal
Next Materials
Published
2026-11-20
DOI
https://doi.org/10.1016/j.nxmate.2026.104079
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Hybrid machine learning framework for the classification and prediction of dry sliding wear performance of friction stir processed and laser shock peened AA6061-O alloy

Sridharan Mohan, Nowsath Begam Mohammed Yusuf, Balasundaram Rathinam
Next Materials
Machine Learning in Materials Science
article

Hybrid machine learning framework for the classification and prediction of dry sliding wear performance of friction stir processed and laser shock peened AA6061-O alloy

Sridharan Mohan, Nowsath Begam Mohammed Yusuf, Balasundaram Rathinam
article en

Abstract

This study presents a hybrid machine learning model (classification and regression) based framework to understand the tribological properties of Aluminium alloy 6061 (AA6061). At first hybrid experimental approach combining friction stir processing (FSP) and laser shock peening (LSP) was conducted with the AA6061-O alloy (base material). For this, wear rate (W R ) of AA6061 is observed as a function of variable applied load (AL - 10, 20, and 30 N), variable sliding velocity (SV - 1, 2, and 3 m/s) and constant sliding distance (SD – 1000 m). Analysis of variance (ANOVA) showed that AL and SV significantly influenced W R (p < 0.05), contributing more than 89% to the total variation. Results obtained from the real-time experiments are then used to train, test and validate both machine learning (ML) models. Results recorded by the J48 decision tree proven its balanced classifying ability across both HIGH and LOW classes. The weighted average F-measure of 0.892 for BM, 0.886 for FSP, and 0.886 for LSP. Meanwhile, results recorded by the extreme gradient boosting (XGBoost) validated its combined gradient boosting nature with regularization and effective handling of nonlinear relationships, and strong generalization. Also, this study attempts the generalization of the proposed model through the model interpretability study. Results of interpretability analysis confirms SV as the first and AL as the second order influential parameters in governing the wear rate. Thus, the proposed hybrid framework offers a reliable and effective tool for understanding the tribological behavior of AA6061.

Next MaterialsVol. 14
SRM Institute of Science and Technology (IN)
Institution of Engineers (India)
Machine Learning in Materials Science
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon