Scribd
Upload
281 views21 pages

AI and Machine Learning For NDT

The document discusses applying AI and machine learning to non-destructive testing (NDT). It outlines the business case for adopting technology in the NDT field due to issues like an aging workforce and rising labor costs. Examples of AI applications in other industries are provided. The document then describes three case studies of AI/ML being applied to NDT tasks like predicting component life, automating ultrasonic data analysis, and automated defect detection in digital radiography. Barriers to adoption in NDT are discussed as well as next steps like accessing large, standardized datasets and starting with small proof-of-concept projects.

Uploaded by

Alexander Tirado
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
281 views21 pages

AI and Machine Learning For NDT

The document discusses applying AI and machine learning to non-destructive testing (NDT). It outlines the business case for adopting technology in the NDT field due to issues like an aging workforce and rising labor costs. Examples of AI applications in other industries are provided. The document then describes three case studies of AI/ML being applied to NDT tasks like predicting component life, automating ultrasonic data analysis, and automated defect detection in digital radiography. Barriers to adoption in NDT are discussed as well as next steps like accessing large, standardized datasets and starting with small proof-of-concept projects.

Uploaded by

Alexander Tirado
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

AI and Machine Learning for NDT

NDTMA Conference
February 2019
Presentation Outline

• AI definition
• Examples and Notable applications in the
marketplace today
• Application in the NDT field
• Barriers
• Conclusion
• Next Steps
The Business Case for Embracing Technology
in the NDT Community

The case for why this is important to the NDT community:


• NDT service market, globally and USA: $8.5B / $3.4B respectively
• NDT technical staffing level in USA (in-house and out sourced) is
26,000. Predicted to rise to 33,000 by 2023
• Aging workforce issue is particularly evident in the NDT field
• Industry margins will be challenged due to labor cost inflation
• One industry report documents labor as a percent of revenue has grown from
46% in 2011 to 51% in 2017

• And finally:
• Business valuation multiples for scalable, technology based industrial firms are
7 – 10 X vs NDT service firm multiples in the 5 – 7 X EBITDA range
AI / ML / DL – Which is it ?
Some Quick Examples
Elements of Artificial Intelligence
Common Applications for AI
Typical Applications

Let’s start with the most obvious …


But Applications
Cross All Industries
• Autonomous driving
• Chatbots / Customer
service
• HR / Recruiting
Automation

• Pharmaceutical
• Journalism
• Household appliances
NDT Case Studies
AI / ML Case Study #1
Workpad / TechKnowServ team

• Goal: Predict component remaining life


• Participants:
• TechKnowServ (NDT service provider)
• Workpad, LLC (Technology enabler)

• Inspection framework:
• RFID tagged assets
• Danatronics Ultrasonic thickness data collection with wireless data transfer to Cloud
environment
• Web application for display of thickness data

• Integration of Machine Learning


• Workpad integrated Amazon Machine Learning Service
• Use ML algorithms to discover patterns in data and construct predictive models.
AI / ML Case Study #2
Ultrasonic Phased Array Data Analysis

• Goal: Reduce PA data analysis time / Improve consistency of


analysis
• Participants
• Veriphase
• Regional NDT service provider
• Inspection framework:
• Typical PAUT data collection with Olympus OmniPC
• Veriphase ADT program for data processing

• Integration of Machine Learning


• Rules built around 27 essential variables
• Proprietary algorithms for geometry classification.
AI / ML Case Study #2
Ultrasonic Phased Array Data Analysis

Key Benefits:
• Machine Learning protocols allow ADT to:
• Serve as the initial screening / decision tool
• Allow ADT to be used as a training tool
• Can be used as a QA data check
• Achieve 90% reduction in analysis time.
• Software will allow PA data analysis at a rate faster than
acquisition rate.
• Real world example:
• 30 – 45 min to acquire 500 mb of PA data in a 20’ weld
• Less than one minute to analyze and serve up only relevant indications
Digital Radiography #3

• Goal: Automated discontinuity detection


• Participants:
• BAM Federal Institute (Berlin Germany)
• Dept of Computer Science, Pontifical Catholic University of Chile

• Inspection framework:
• Collection of 19,400 radiographs
• Digitized from conventional acetate film

• Integration of Machine Learning


• University of Chile developed large scale, image database.
• Formed GD Xray to manage commercially accessible database of weld and casting
images.
Medical / Industrial Parallel
Today, one of the biggest problems facing physicians and clinicians in general is the
overload of too much patient information to sift through. This rapid accumulation of electronic data is
thanks to the advent of electronic medical records (EMRs) and the capture of all sorts of data about a patient that was not previously
recorded, or at least not easily data mined. This includes imaging data, exam and procedure reports, lab values, pathology reports,
waveforms, data automatically downloaded from implantable electrophysiology devices , data transferred from the
imaging and diagnostics systems themselves, as well as the information entered in the EMR, admission, discharge and
transfer (ADT), hospital information system (HIS) and billing software. In the next couple years there will be a further data explosion with the
use of bidirectional patient portals, where patients can upload their own data and images to their EMRs. This will include images shot with
their phones of things like wound site healing to reduce the need for in-person follow-up office visits. It also will include medication
compliance tracking, blood pressure and weight logs, blood sugar, anticoagulant INR and other home monitoring test results, and activity
the evolving Internet of things (IoT) to aid in keeping patients
tracking from apps, wearables and

healthy. Physicians liken all this data to drinking from a firehose because it is overwhelming. Many say it is very difficult or
impossible to go through the large volumes of data to pick out what is clinically
relevant or actionable. It is easy for things to fall through the cracks or for things to be lost to patient follow-up. This issue is
further compounded when you add factors like increasing patient volumes, lower reimbursements, bundled payments and the conversion
from fee-for-service to a fee-for-value reimbursement system. This is where artificial intelligence will play a key role in the next couple years.
AI will not be diagnosing patients and replacing doctors — it will be augmenting
their ability to find the key, relevant data they need to care for a patient and present
it in a concise, easily digestible format.
Excerpt from Image Technology news: “How Artificial Intelligence Will Change Medical Imaging”, Feb, 24, 2017
AI Adoption Barriers
• No industry effort to coordinate data standardization
• Lack of centralized, open source databases for training the
computer / AI system
• Lack of clear protocol for labeling data (i.e. DICONDE for digital
radiography)
• Labor hour driven service model provides a headwind.
Conclusions and Next Steps
• AI applications are mainstream in adjacent industries
• AI and ML application is nascent in Industrial NDT
– Lack of sufficient scale opportunity
• Commercially available tools like Amazon Machine Learning
service, and OpenVDNA Project1 make technology accessible at
low cost
• Key issue is access to normalized and large datasets
– Role for NDTMA / ASNT ??
• Engage external consultants that have proven success
• Start with small, quick wins to secure organizational buy in.
1 – Ref. NDT.ORG, 10-31-18
Questions ?
Appendix
OpenVDNA Project
• Exascale Visual AI initiative to sequence Visual DNA from images
using Volume Learning to enable new applications.
• The OpenVDNA project provides public resources for visual AI
solutions, including a visual DNA catalog, visual AI search engine,
and a visual AI toolkit.
• Visual DNA are taken from thousands of small pieces of images
composing the fabric of the image.
• Visual DNA elements are analyzed and compared using over
16,000 feature metrics in four primary bases: Color, Shape,
Texture and Glyphs.
• Visual DNA are learned and stored in the OpenVDNA catalog
using Learning Agents to create structures representing higher-
level objects, and then accessed via the OpenVDNA search engine
and the OpenVDNA software toolkit.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy