Definition
The estimated rate at which a signal classified as unwanted by an alarm verification model was in fact an actual event, estimated on a deliberately dispatched audit sample because the suppressed population produces no confirming record.
The Measurement Gap
There is no metric that tells you how often a signal classified as unwanted by an alarm verification model was in fact an actual event, which means nuisance alarm reduction is reported as an unqualified success and the suppression loss it produces never enters any record.
Formula
VSLR = (r_audit / n_audit) × N_suppressed , reported as an estimated count with a confidence interval, and as a rate r_audit / n_audit
Proposed formula, pending validation pilots.
At a Glance
Decision Supported
Whether nuisance alarm reduction can be expanded, held, or must be rolled back, and at what verification threshold.