where expertise comes together - since 1996

Web's Largest Portal of Nondestructive Testing (NDT)
Open Access Database (Conference Proceedings, Articles, News), Exhibition, Forum, Network

All Forum Boards
Technical Discussions >
Probability of detection
Career Discussions
Job Offers
Job Seeks
Classified Ads
About NDT.net
Articles & News

Sonatest Ltd
From our three distinct business centres in the UK, USA and CANADA, Sonatest design and produce a leading range of high performance ultrasonic NDT equipment and accessories.

961 views
06:19 Apr-16-2002

Wang

Consultant
United Kingdom,
Joined Apr 2002
2
Probability of detection

Dear All

I am currently doing a literature survey to try to derive some meaningful generic format for POD (Probability of detection), to welded structures for a start. We are aiming to have a standard procedure at the end. This is following some large projects in the European scene, e.g. PISC, TIP, NORDTEST etc. It is noted that the probability of detection is dependent on so many points but to date, it is still an art rather than science, despite considerable investment over the many years.

I would like to know, 1) any new approach to treat this old problem
2) any relevant literature, e.g. data bank, QNDE etc.

With many regards




 
03:55 Apr-17-2002

Dave

R & D
TRI/Austin,
USA,
Joined Nov 2001
41
Re: Probability of detection Dear Wang:

It is not clear to me what you mean by a "format for POD".

If you are looking for references describing how to conduct experiments to determine POD, I suggest you read:

Floyd Spencer, Giancarlo Borgonovi, Dennis Roach, Don Schurman, Ron Smith
Reliability Assessment at Airline Inspection Facilities Volume I:
A Generic Protocol for Inspection Reliability Experiments
DOT/FAA/CT-92/12, I
March 1993

Nondestructive Evaluation System Reliability Assessment
MIL-HDBK-1823
30 April 1999
(also available previously as AGARD-LS-190)

These are of course aerospace-based.

For a collection of POD data, see the NDE Capabilities Databook, 3rd edition, go to http://www.ntiac.com

As for it being an art, I don't think that is the case. The experimental methods and statistical analysis methods for the experiment results are mature and have been published.

What do you desire from a new approach?

Regards, Dave.

david.forsyth@nrc.ca

: Dear All
.
: I am currently doing a literature survey to try to derive some meaningful generic format for POD (Probability of detection), to welded structures for a start. We are aiming to have a standard procedure at the end. This is following some large projects in the European scene, e.g. PISC, TIP, NORDTEST etc. It is noted that the probability of detection is dependent on so many points but to date, it is still an art rather than science, despite considerable investment over the many years.
.
: I would like to know, 1) any new approach to treat this old problem
: 2) any relevant literature, e.g. data bank, QNDE etc.
.
: With many regards
.



 
03:37 Apr-19-2002

Terry Oldberg

Engineering, Mechanical Electrical Nuclear Software
Consultant,
USA,
Joined Oct 1999
42
Re: Probability of detection I have to disagree with Dave when he suggests that the process is well established, for the paper "Erratic Measure" (see http://www.ndt.net/article/v04n05/oldberg/oldberg.htm ) demonstrates that attempts at establishing a process for the statistics of defect detection have failed and that this failure is inherent in the process of defect detection itself. Thus, while advice on how to conduct studies on the statistics of NDT is indeed available, this advise must be incorrect.

: Dear Wang:
.
: It is not clear to me what you mean by a "format for POD".
.
: If you are looking for references describing how to conduct experiments to determine POD, I suggest you read:
.
: Floyd Spencer, Giancarlo Borgonovi, Dennis Roach, Don Schurman, Ron Smith
: Reliability Assessment at Airline Inspection Facilities Volume I:
: A Generic Protocol for Inspection Reliability Experiments
: DOT/FAA/CT-92/12, I
: March 1993
.
: Nondestructive Evaluation System Reliability Assessment
: MIL-HDBK-1823
: 30 April 1999
: (also available previously as AGARD-LS-190)
.
: These are of course aerospace-based.
.
: For a collection of POD data, see the NDE Capabilities Databook, 3rd edition, go to http://www.ntiac.com
.
: As for it being an art, I don't think that is the case. The experimental methods and statistical analysis methods for the experiment results are mature and have been published.
.
: What do you desire from a new approach?
.
: Regards, Dave.
.
: david.forsyth@nrc.ca
.
: : Dear All
: .
: : I am currently doing a literature survey to try to derive some meaningful generic format for POD (Probability of detection), to welded structures for a start. We are aiming to have a standard procedure at the end. This is following some large projects in the European scene, e.g. PISC, TIP, NORDTEST etc. It is noted that the probability of detection is dependent on so many points but to date, it is still an art rather than science, despite considerable investment over the many years.
: .
: : I would like to know, 1) any new approach to treat this old problem
: : 2) any relevant literature, e.g. data bank, QNDE etc.
: .
: : With many regards
: .
.



 
03:46 Apr-26-2002

Tim MacInnis

Senior NDT Engineer
SAIC/Ultra Iamge International,
USA,
Joined Dec 1999
10
Re: Probability of detection There are two publications I work from regarding PODs.
The first is the DOD/FAA pub "Reliability Assessment at Airline Inspection Facilities, V1: A Generic Protocol for Inspection Reliabilty Experiemnts", 1992.
The second publication is the DOD Hanbook "Nondestructive Evaluation System Reliability Assessment MIL-HNBK-1823, 30 Aprol 1999.
You may find the "format" you are looking for in these very informative publications.


 
02:53 May-06-2002

Terry Oldberg

Engineering, Mechanical Electrical Nuclear Software
Consultant,
USA,
Joined Oct 1999
42
Re: Probability of detection Though I've not read the referenced documents, I doubt that they provide a solution for Wang. Documents of their era ignore a foundational weakness that is exposed in my 1995 paper, "Erratic Measure" ( http://www.ndt.net/article/v04n05/oldberg/oldberg.htm ). The weakness that I reference is violation of the axiom of probability theory that is known as "Unit Measure."

A false doctrine of this era makes the POD the measure of the reliability of a type of test that violates Unit Measure. If I'm not mistaken, the documents that you reference embrace this doctrine.

I do know that a MIL-SPEC that was in preparation by the DOD, circa 1990, embraced this doctrine and suspect that MIL-HNBK-1823 is this same document. I also know that the FAA's post-1992 research on Aging Aircraft fails to preserve Unit Measure and that the publication date of the FAA/DOD document that you reference is 1992.






 


© NDT.net - The Web's Largest Portal of Nondestructive Testing (NDT) ISSN 1435-4934

Open Access Database, |Conference Proceedings| |Articles| |News| |Exhibition| |Forum| |Professional Network|