fpf.tools
Library / Systems Engineering Principles Framework
Jump to passage
In this reading
Source · lines 12035–12051

Link to current text

Published source confirmed at last check

Source changed 2026-10-03 08:25:59 UTC · snapshot created 2026-10-03 08:26:43 UTC · last check 2026-10-03 09:15:10 UTC

SYSE.46:5.1 - Separate support benefit from the choice to obtain it

In a constructed service-update diagnostic, the optional contribution is documentation lookup; execution, current-state checks and permission remain common. E supplies an applicable procedure premise, R leaves a decisive procedure fact only in the reachable source, and T supplies a sufficient premise while offering an irrelevant source.

SituationNo optional lookupSource operation suppliedAgent chooses support under the current rule
E: sufficient applicable premiseCompletes correctly.Completes, with acquisition that adds no premise.Completes after two redundant lookups.
R: decisive premise missingReturns the missing fact; completion is unsupported.Binds and performs the source call, uses the fact and completes.Skips the available call, guesses and fails.
T: sufficient premise, irrelevant offered sourceCompletes correctly.Rejects the irrelevant return and completes from its premise.Repeats the irrelevant lookup twice, then completes from the premise.

The stipulated R trace locates missed assistance selection, because the usable call was skipped. If the call had been chosen with a wrong target, repair invocation instead; if the returned fact disappeared from the next input, repair input preparation; if it remained visible but control sent the task back to retrieval, repair the procedure. E and T reveal unnecessary work despite correct answers. Count required effect verification and a sufficient-result stop as well as calls and completion.

After the selected rule repair, unused matched cases must show E using its supplied premise, R obtaining and using the needed source, and T declining irrelevant acquisition, while keeping the actual-state and at-most-one-effect checks. Fewer calls with a guessed R answer fails. The Reference works the whole repair comparison. This table defines an illustrative diagnostic, not observed effectiveness.