French (Fr)English (United Kingdom)

DPC NEWS: a website dedicated to Penetrant Testing and Magnetic Testing



visits on site since April 2008

Log in


Receive HTML?

March-May 2015 :NDT documents: if two signatures were enough!

Written by Laurence
Thursday, 19 February 2015 12:14

The attention of one of us was drawn about conversations on LinkedIn emanating from Filitsa Tileli, NDT Engineer at Strata, as Top Contributor.

Here are the contents:

“Good morning people,

This discussion subject was inspired by a comment that James Stoney did in another discussion about time wasters.

You all know, no matter in which sector of industry you are occupied that documentation needs to be signed and stamped in order to be issued or considered valid. Whether it has to do with NDT reports, written instructions, procedures, purchase orders, training sheets, exam papers, lab reports, you name it, it needs a signature. And in most cases, a stamp as well.

I have been in companies where the NDT work instructions need to be approved by the manufacturing engineers and/or quality engineers. In some cases, even by program managers. I have been in situations where I needed 4 signatures on top of the author of the document and the responsible Level III. The rest of these signatures were to come from people that know about NDT as much I know about perfume making. And although I may be able to appreciate a good smell, I still know bugger all about the process and I could not possibly sign any document related to the making of it.

So, I have found myself in cases where I chase people for signatures who know nothing about NDT and, being weary of signing anything that they don't understand, ask for time to review the procedure. Then they come across things they do not understand and want explanations and I have to run a crash course in NDT just to get that signature and move to the next one.

How wise is to need approval signatures from people that do not know anything about NDT? After all NDT is a special process and you have qualified people to do the review. I would think that at most you need just one more signature apart from the responsible Level III from a manager which would just serve as a verification that this person was notified that a new NDT document is out.

Associated to this, there is the larger problem of who is responsible to provide authorization for which area or which subject. I can be asked to review a manufacturing work instruction or procedure in order to verify that the NDT inspection is in the right place but do I need to officially sign a manufacturing work instruction?

Which documents need reviewing by which people? Do the extra signatures add more weight to the document or do they serve in pacifying our own sense of insecurity? In the sense that we keep adding more signatures and reviews to a document, hoping we are creating a failsafe and that in case of a problem the responsible for it can be found?

Do we actually create more work and eventually more confusion in case of a failure? Because if a failure occurs what is the use of finding out, upon investigation, that half the people who signed a document may had as well signed a purchase order of toilet paper for all the good it did?

I do agree that we need reviews and check points to make sure that we do not let problems into a system. But which changes need signatures by which people? If I change the code number or the issue of a document mentioned in a procedure, do I need all the array of signatures? If the procedure is a new one would I need new more signatures by more people i.e. more reviews since it is a new document? Document traceability is important but how do I maintain the importance without creating ‘snowdrifts’ that make the actual traceability recording more difficult? Is it really worth to go on doing it this way?

I think that linked to this is the red tape problems that bug the aerospace industry as well. But I would like to have that in a separate discussion as it is also linked to availability of information in order for tasks to be completed.

So, all you military aviation guys that already got twichy (is that a Pavlovian response? Mention to military guys the word “security” and they do start to get twichy!!) calm down for the time being, I promise I will set this up as a separate discussion.”


Here are our comments:

Who does sign the specifications?

For a more complete picture, we performed a small investigation by examining some European and American specifications, in a relatively limited in number, from the aerospace and nuclear sectors. It appears that some specifications include:

  • Only one signature: either that of a person who, alone, has the authority to sign a specification that has been validated by all, or that of the writer who is also the person who approves the document;
  • Two signatures: those of the drafter and of the auditor or of the person who approves the document;
  • Four signatures: the writer; as well as the auditor, the person who approves and, incredibly, ...a so-called Additional visa!


What should we think?

  • Only one signature. When the writer is also the person who approves the document, we may think that no one else has checked the specification; as "nobody's perfect", there may be errors; this is what we found at least once;
  • Two signatures: those of the writer and of the auditor or of the person who approves the document. This seems enough;
  • Four signatures: why so many? We think it is too much. Why an auditor and a person who approves? What may we think about the "additional visa" whose job title is not specified?
  • While one signature i.e writer/approver is not enough, two seem to do the job for low level technical documents and work instructions. Three signatures could be utilized in higher level documents with the third signature serving as confirmation that a manager or director has received notice that a new procedure is out. For example, a work instruction for the inspection of particular parts in the shop floor could have the signature of the author (usually a Level II NDT inspector or NDT engineer) with the approval signature of the Level III validating the document and leading to its being issued on the system. A higher level document, a procedure dealing with a certain NDT method would need the signature of the author, the responsible Level III and the manager who runs the section.The document could be reviewed by as many people as one wants, in order to get feedback for it (a good practice is to give the document out to someone not related to NDT, if only to make sure that the text makes sense. If that person can get the general idea, then you have a proof for the coherence of the document!) but not all these people need to sign the document. Review and authorization should be two different things.


Signing the documents

Regarding the signing of documents, four concepts are to be considered:

  • The immediate supervisor, "jealous of his authority," who wants to "see everything, sign everything;"
  • The immediate supervisor, who does not know how to delegate, that is to say, who does not trust his subordinates. This fear may be justified if the subordinate is not "reliable." However, is this not the fault of the person who hired or appointed some unqualified person for such a position? Often, this occurs in very small and small companies that originally have a Director who does everything, thus, holds several positions of responsibility. Gradually, as the company grew, he created manager positions, but he cannot get rid of the idea of "seeing everything, signing everything." If the boss does not change his managing way, he will scatter and will have no time enough to dedicate to its true job: that of the boss who must ensure the development and the long-term success of his company. However, we must not underestimate the general civil and criminal liability of the Manager who may be sentenced to jail for delinquency;
  • The immediate supervisor who knows how to delegate and has confidence in his team, which is certainly and fortunately the most common case.
  • The document control/quality people that ask for more signatures in the high level document control procedures. Usually these high level procedures are drawn by people that want to put as many fail safes as possible. At this stage what is confused is the review and authorization of a document. As was stated before, a document can be reviewed by a lot of people. However, the authorization need only depends on a small number of people. As an example, an NDT work instruction can be circulated to all operators in the draft form and the comments fed back to the author for necessary corrections. Not all these people need to sign the document, though. The same principle can be applied to the rest of the signatures that are needed.


Each of us has witnessed, during his professional career, "incredible" signatures situations.

Such is the example of the manager who, leaving lately for an appointment or meeting, is "chased" and in extremis caught again by a member of his department who wants him to sign an urgent document. Thus, the manager lays down his briefcase and signs the paper on the corridor wall, or in any other acrobatic situation. The problem is that the manager does not have, generally, time enough to read anything, which does not prevent him from signing and "flying" to his appointment or meeting.

Another case: a person does not want to assume his/her responsibilities, and wants to "cover himself/herself", and has his/her immediate supervisor sign the document.

The main problem seems to be the confusion between review and authorization. One should think of the compilation of international standards. These documents are available to the public in their draft form and there is a period of the review, during which professionals that will use the standard are free to comment and offer suggestions. Once this period is over the comments are examined and necessary modifications are made to the standard which will then be published. None of the commentators’ signatures appears on these standards or the signature pages would be more than the standard pages.

So, why can’t the companies sent out documents for review and ask for comments to be sent back to the author? This way the author can incorporate any changes needed and the final draft can go straight for the final checking and approval of the person who has that authority. If the review and feedback stage is kept separate, then the amount of signatures needed on the front page of a document can be reduced drastically.

Will it be so? Hopefully more companies will realize that and start moving in the right direction.

Last Updated ( Thursday, 19 February 2015 12:29 )