Distinguish a Failed Measurement from an Error or Skipped Test Step

A test station must distinguish a valid measurement outside its limit from a measurement that never completed. It should also distinguish an intentional skip from a passing check. Collapsing all of these outcomes into one red or green indicator makes production decisions and failure analysis unreliable.

NI’s TestStand status-expression explanation identifies separate statuses including passed, failed, skipped, done, and error. Your application may use another vocabulary, but the underlying distinctions should remain explicit in the saved result.

Define two kinds of outcome

First record whether the operation executed successfully and produced valid evidence. Then record what that evidence says about the product requirement. A communication timeout may prevent a voltage test from completing; it does not, by itself, establish that the board’s voltage was outside specification.

Give every required step a stable identifier and an expected evidence type. For a numerical test, that may include value, unit, limits, range, and relevant acquisition settings. For a visual or operator check, define what observation must be recorded and who may make the disposition.

Specify the unit-level decision

Do not make a missing measurement equivalent to zero, because zero may be a valid reading. Store missingness and error context explicitly. Likewise, a script returning without an exception does not prove that it performed the intended test. If an optional step becomes mandatory in another variant, verify both configurations; the same skip status may require a different unit disposition.

Verify abnormal paths

1. Simulate an instrument communication failure.
2. Exercise an intentional skip and a variant-specific omission.
3. Check how interrupted execution affects the overall result.
4. Confirm the saved report preserves the distinction.
5. Verify that required incomplete testing cannot become an unexplained pass.

Use controlled test environments for these checks and restore the approved configuration afterward. Review cleanup behavior as well, including whether the station leaves outputs or relays in the intended state after an interruption.

Close the reporting design with a state map that operators and quality reviewers can understand. Preserve raw evidence when a valid measurement exists, and preserve error context when it does not. The retest history guide explains how later attempts should relate to these earlier outcomes.