A condition in which two organizations intentionally do not share some particular set of goals. Administrative Independence is usually established to at least partially substantiate that one of the two organizations can plausibly provide cognitively unbiased witnessing of execution by the other.
The witnessing group is often rigidly process-oriented, with many intermediate opportunities to document the execution of their tasks. In most cases, the witnessing group provides no more than very limited interpretation1 and, therefore, requires the executing group to provide explicit criteria to be witnessed. The relationship is conceptually similar to that provided by a Notary Public with respect to a signature. The notary does not interpret the document signed: they only verify that that the signer provided sufficient verification of identity for their signature to be credible.
Administrative independence is actually a matter of degree: the two groups in the pair will often have some intersection of management. The more separation of budgetary and objective management that exists between them, the more reasonable it is to expect true objectivity. In most cases (however) there is also an overlay of certification by one or more customer organizations2. In some situations, the customer oversight defines an explicit set of competing motivations, some of which can impose severe penalties on the witnessing group in the event of corruption.
Examples of this relationship include Quality Assurance (QA) with Manufacturing, QA with Test, Configuration Management with Engineering, Test with Engineering, and Authorized Representatives with pretty much everything else.
Footnotes- If any at all.[↩]
- Which might be either Acquisition or Regulatory.[↩]