From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy

A method is developed to exploit data on complex materials behaviors that are impossible to tackle by conventional machine learning tools. A pairwise comparison algorithm is used to assess a particular property among a group of different alloys tested simultaneously in identical conditions. Even tho...

Full description

Saved in:
Bibliographic Details
Main Authors: Rafael Herschberg, Lisa Rateau, Laure Martinelli, Fanny Balbaud-Célérier, Jean Dhers, Anna Fraczkiewicz, Gérard Ramstein, Franck Tancret
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Metals
Subjects:
Online Access:https://www.mdpi.com/2075-4701/14/12/1412
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850085090916302848
author Rafael Herschberg
Lisa Rateau
Laure Martinelli
Fanny Balbaud-Célérier
Jean Dhers
Anna Fraczkiewicz
Gérard Ramstein
Franck Tancret
author_facet Rafael Herschberg
Lisa Rateau
Laure Martinelli
Fanny Balbaud-Célérier
Jean Dhers
Anna Fraczkiewicz
Gérard Ramstein
Franck Tancret
author_sort Rafael Herschberg
collection DOAJ
description A method is developed to exploit data on complex materials behaviors that are impossible to tackle by conventional machine learning tools. A pairwise comparison algorithm is used to assess a particular property among a group of different alloys tested simultaneously in identical conditions. Even though such characteristics can be evaluated differently across teams, if a series of the same alloys are analyzed among two or more studies, it is feasible to infer an overall ranking among materials. The obtained ranking is later fitted with respect to the alloy’s composition by a Gaussian process. The predictive power of the method is demonstrated in the case of the resistance of metallic materials to molten salt corrosion and wear. In this case, the method is applied to the design of wear-resistant hard-facing alloys by also associating it with a combinatorial optimization of their composition by a multi-objective genetic algorithm. New alloys are selected and fabricated, and their experimental behavior is compared to that of concurrent materials. This generic method can therefore be applied to model other complex material properties—such as environmental resistance, contact properties, or processability—and to design alloys with improved performance.
format Article
id doaj-art-9e40962bbdc244ee9dbdb146936b91bf
institution DOAJ
issn 2075-4701
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Metals
spelling doaj-art-9e40962bbdc244ee9dbdb146936b91bf2025-08-20T02:43:49ZengMDPI AGMetals2075-47012024-12-011412141210.3390/met14121412From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design StrategyRafael Herschberg0Lisa Rateau1Laure Martinelli2Fanny Balbaud-Célérier3Jean Dhers4Anna Fraczkiewicz5Gérard Ramstein6Franck Tancret7Nantes Université, CNRS, Institut des Matériaux de Nantes Jean Rouxel, IMN, 44000 Nantes, FranceNantes Université, CNRS, Institut des Matériaux de Nantes Jean Rouxel, IMN, 44000 Nantes, FranceCEA, Service de Recherche en Corrosion et Comportement des Matériaux, Université Paris-Saclay, 91191 Gif-sur-Yvette, FranceCEA, Service de Recherche en Corrosion et Comportement des Matériaux, Université Paris-Saclay, 91191 Gif-sur-Yvette, FranceFramatome, 69006 Lyon, FranceMINES Saint-Etienne, Université de Lyon, CNRS, UMR 5307 LGF, Centre SMS, 42023 Saint-Etienne, FranceNantes Université, CNRS, Laboratoire des Sciences du Numérique de Nantes (LS2N), 44306 Nantes, FranceNantes Université, CNRS, Institut des Matériaux de Nantes Jean Rouxel, IMN, 44000 Nantes, FranceA method is developed to exploit data on complex materials behaviors that are impossible to tackle by conventional machine learning tools. A pairwise comparison algorithm is used to assess a particular property among a group of different alloys tested simultaneously in identical conditions. Even though such characteristics can be evaluated differently across teams, if a series of the same alloys are analyzed among two or more studies, it is feasible to infer an overall ranking among materials. The obtained ranking is later fitted with respect to the alloy’s composition by a Gaussian process. The predictive power of the method is demonstrated in the case of the resistance of metallic materials to molten salt corrosion and wear. In this case, the method is applied to the design of wear-resistant hard-facing alloys by also associating it with a combinatorial optimization of their composition by a multi-objective genetic algorithm. New alloys are selected and fabricated, and their experimental behavior is compared to that of concurrent materials. This generic method can therefore be applied to model other complex material properties—such as environmental resistance, contact properties, or processability—and to design alloys with improved performance.https://www.mdpi.com/2075-4701/14/12/1412alloy designmachine learningoptimizationmolten salt corrosionwear
spellingShingle Rafael Herschberg
Lisa Rateau
Laure Martinelli
Fanny Balbaud-Célérier
Jean Dhers
Anna Fraczkiewicz
Gérard Ramstein
Franck Tancret
From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
Metals
alloy design
machine learning
optimization
molten salt corrosion
wear
title From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
title_full From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
title_fullStr From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
title_full_unstemmed From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
title_short From Pairwise Comparisons of Complex Behavior to an Overall Performance Rank: A Novel Alloy Design Strategy
title_sort from pairwise comparisons of complex behavior to an overall performance rank a novel alloy design strategy
topic alloy design
machine learning
optimization
molten salt corrosion
wear
url https://www.mdpi.com/2075-4701/14/12/1412
work_keys_str_mv AT rafaelherschberg frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT lisarateau frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT lauremartinelli frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT fannybalbaudcelerier frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT jeandhers frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT annafraczkiewicz frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT gerardramstein frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy
AT francktancret frompairwisecomparisonsofcomplexbehaviortoanoverallperformancerankanovelalloydesignstrategy