August 2025
Volume 83 • Number 8
JOURNAL STAFF
PUBLISHER: Neal J. Couture, CAE
DIRECTOR OF PUBLICATIONS/
EDITOR: Jill Ross
ASSOCIATE EDITOR:
Stefanie Laufersweiler
PRODUCTION MANAGER: Joy Grimm
DIGITAL PUBLISHING MANAGER:
Synthia Jester
DIGITAL CONTENT STRATEGIST:
Haley Cowans
ASNT MEDIA &EVENT SALES
Peter Roy, proy@asnt.org
1-614-384-2431
Sonny Hines, shines@asnt.org
1-614-384-2434
TECHNICAL EDITOR
John Z. Chen, KBR
ASSOCIATE TECHNICAL EDITORS
John C. Aldrin, Computational Tools
Sreenivas Alampalli, Stantec
Ali Abdul-Aziz, Kent State University
Yiming Deng, Michigan
State University
Dave Farson, Ohio State University
Jin-Yeon Kim, Georgia
Institute of Technology
Mani Mina, Iowa State University
Ehsan Dehghan-Niri,
Arizona State University
Yi-Cheng (Peter) Pan, Emerson Inc.
Anish Poudel, MxV Rail
Donald J. Roth, Roth
Technical Consulting LLC
Ram P. Samy, Consultant
Steven M. Shepard,
Thermal Wave Imaging
Ripi Singh, Inspiring Next
Surendra Singh, Honeywell
Roderic K. Stanley, NDE
Information Consultants
Matthew Webster, NASA
Langley Research Center
Lianxiang Yang, Oakland University
Reza Zoughi, Iowa State University
CONTRIBUTING EDITORS
Toni Bailey, TB3NDT Consulting
Megan McGovern,
General Motors Corp.
Saptarshi Mukherjee, Lawrence
Livermore National Laboratory
Hossein Taheri, Georgia
Southern University
UPFRONT
|
SCANNER
UNCERTAINTY
QUANTIFICATION AND
ANALYSIS IN NDE
Uncertainty quantification (UQ) and analysis are vital components in the
advancement and reliability assurance of nondestructive evaluation (NDE)
methodologies. This special issue of Materials Evaluation presents cutting-
edge research, providing valuable insights and novel approaches to
addressing uncertainties inherent in NDE processes across aerospace, infra-
structure, and other critical industries.
The tutorial article by Prof. Zi Li and myself offers a comprehensive review
and introduces a refined framework for categorizing and mitigating uncer-
tainties in NDE. We propose an advanced structure that goes beyond tradi-
tional aleatoric and epistemic classifications by detailing uncertainties specific
to data, forward modeling, and inverse learning. Our analysis emphasizes
the importance of integrating probabilistic, statistical, simulation-based, and
AI-driven methodologies, particularly highlighting recent developments such
as digital twins and real-time autonomous inspection systems.
Dr. John Aldrin’s contribution significantly advances model-based inver-
sion techniques, particularly for crack sizing in multilayer fastener sites using
bolt-hole eddy current methods. His research not only develops and vali-
dates robust inversion models but also presents practical guidelines for
assessing the critical 95% safety limit against the uncertainty in undersizing,
which is crucial for ensuring structural integrity using UQ methodologies.
In their insightful paper, Dr. Christine Knott and her coauthors expand
the boundaries of traditional probability-of-detection (POD) analyses.
Recognizing the critical impact of additional variables such as material types,
defect categories, instrumentation settings, and inspector variability, their
work provides a systematic method for determining which additional vari-
ables meaningfully contribute to POD. This rigorous, mathematically sound
alternative to traditional transfer function methods enhances the reliability of
POD-based UQ studies and practical inspection decision-making.
The final paper by Dr. Noritaka Yusa and Dr. Takuma Tomizawa addresses
the challenge of accurately sizing fatigue cracks using measured eddy current
signals. Their probabilistic inversion algorithm effectively quantifies not only the
estimated crack dimensions but also the associated measurement uncertain-
ties. Their work demonstrates the value of probabilistic methods in improving
the reliability of defect characterization in operational environments.
Together, these articles underscore the importance and breadth of uncer-
tainty quantification and analysis in NDE. They offer robust solutions for
integrating theoretical and applied approaches, from advanced statistical
models and inversion techniques to sophisticated, probabilistic AI-driven
methods. These contributions are poised to enhance confidence in inspec-
tion outcomes, significantly reduce false positives and negatives, and foster
more accurate maintenance and safety decisions.
As guest editor of this issue and on behalf of the ME staff, I extend our
sincere thanks to all contributing authors for their exemplary work and hope
readers will find these articles as inspiring and impactful.
YIMING DENG, ASNT FELLOW
PROFESSOR AND DIRECTOR OF NDE LABORATORY
MICHIGAN STATE UNIVERSITY
DENGYIMI@EGR.MSU.EDU
Together, these
articles offer
robust solutions
for integrating
theoretical
and applied
approaches,
from advanced
statistical models
and inversion
techniques to
sophisticated,
probabilistic
AI-driven
methods.
A U G U S T 2 0 2 5 • M AT E R I A L S E V A L U AT I O N 7
Volume 83 • Number 8
JOURNAL STAFF
PUBLISHER: Neal J. Couture, CAE
DIRECTOR OF PUBLICATIONS/
EDITOR: Jill Ross
ASSOCIATE EDITOR:
Stefanie Laufersweiler
PRODUCTION MANAGER: Joy Grimm
DIGITAL PUBLISHING MANAGER:
Synthia Jester
DIGITAL CONTENT STRATEGIST:
Haley Cowans
ASNT MEDIA &EVENT SALES
Peter Roy, proy@asnt.org
1-614-384-2431
Sonny Hines, shines@asnt.org
1-614-384-2434
TECHNICAL EDITOR
John Z. Chen, KBR
ASSOCIATE TECHNICAL EDITORS
John C. Aldrin, Computational Tools
Sreenivas Alampalli, Stantec
Ali Abdul-Aziz, Kent State University
Yiming Deng, Michigan
State University
Dave Farson, Ohio State University
Jin-Yeon Kim, Georgia
Institute of Technology
Mani Mina, Iowa State University
Ehsan Dehghan-Niri,
Arizona State University
Yi-Cheng (Peter) Pan, Emerson Inc.
Anish Poudel, MxV Rail
Donald J. Roth, Roth
Technical Consulting LLC
Ram P. Samy, Consultant
Steven M. Shepard,
Thermal Wave Imaging
Ripi Singh, Inspiring Next
Surendra Singh, Honeywell
Roderic K. Stanley, NDE
Information Consultants
Matthew Webster, NASA
Langley Research Center
Lianxiang Yang, Oakland University
Reza Zoughi, Iowa State University
CONTRIBUTING EDITORS
Toni Bailey, TB3NDT Consulting
Megan McGovern,
General Motors Corp.
Saptarshi Mukherjee, Lawrence
Livermore National Laboratory
Hossein Taheri, Georgia
Southern University
UPFRONT
|
SCANNER
UNCERTAINTY
QUANTIFICATION AND
ANALYSIS IN NDE
Uncertainty quantification (UQ) and analysis are vital components in the
advancement and reliability assurance of nondestructive evaluation (NDE)
methodologies. This special issue of Materials Evaluation presents cutting-
edge research, providing valuable insights and novel approaches to
addressing uncertainties inherent in NDE processes across aerospace, infra-
structure, and other critical industries.
The tutorial article by Prof. Zi Li and myself offers a comprehensive review
and introduces a refined framework for categorizing and mitigating uncer-
tainties in NDE. We propose an advanced structure that goes beyond tradi-
tional aleatoric and epistemic classifications by detailing uncertainties specific
to data, forward modeling, and inverse learning. Our analysis emphasizes
the importance of integrating probabilistic, statistical, simulation-based, and
AI-driven methodologies, particularly highlighting recent developments such
as digital twins and real-time autonomous inspection systems.
Dr. John Aldrin’s contribution significantly advances model-based inver-
sion techniques, particularly for crack sizing in multilayer fastener sites using
bolt-hole eddy current methods. His research not only develops and vali-
dates robust inversion models but also presents practical guidelines for
assessing the critical 95% safety limit against the uncertainty in undersizing,
which is crucial for ensuring structural integrity using UQ methodologies.
In their insightful paper, Dr. Christine Knott and her coauthors expand
the boundaries of traditional probability-of-detection (POD) analyses.
Recognizing the critical impact of additional variables such as material types,
defect categories, instrumentation settings, and inspector variability, their
work provides a systematic method for determining which additional vari-
ables meaningfully contribute to POD. This rigorous, mathematically sound
alternative to traditional transfer function methods enhances the reliability of
POD-based UQ studies and practical inspection decision-making.
The final paper by Dr. Noritaka Yusa and Dr. Takuma Tomizawa addresses
the challenge of accurately sizing fatigue cracks using measured eddy current
signals. Their probabilistic inversion algorithm effectively quantifies not only the
estimated crack dimensions but also the associated measurement uncertain-
ties. Their work demonstrates the value of probabilistic methods in improving
the reliability of defect characterization in operational environments.
Together, these articles underscore the importance and breadth of uncer-
tainty quantification and analysis in NDE. They offer robust solutions for
integrating theoretical and applied approaches, from advanced statistical
models and inversion techniques to sophisticated, probabilistic AI-driven
methods. These contributions are poised to enhance confidence in inspec-
tion outcomes, significantly reduce false positives and negatives, and foster
more accurate maintenance and safety decisions.
As guest editor of this issue and on behalf of the ME staff, I extend our
sincere thanks to all contributing authors for their exemplary work and hope
readers will find these articles as inspiring and impactful.
YIMING DENG, ASNT FELLOW
PROFESSOR AND DIRECTOR OF NDE LABORATORY
MICHIGAN STATE UNIVERSITY
DENGYIMI@EGR.MSU.EDU
Together, these
articles offer
robust solutions
for integrating
theoretical
and applied
approaches,
from advanced
statistical models
and inversion
techniques to
sophisticated,
probabilistic
AI-driven
methods.
A U G U S T 2 0 2 5 • M AT E R I A L S E V A L U AT I O N 7
