A carrier moves an ETA by a day, then another two hours. A warehouse resends its file. If every event becomes a separate notification, the real exception disappears in noise. One monitored issue should stay open until resolved.
This workflow requires an enabled logistics module and suitable data sources; check availability in your account.
A practical sequence
Choose a measurable signal: delay against the promised date, sellable-stock cover or age of the last source update.
Set the threshold and check interval. Define when a source is stale; periodic checks are not real-time observation.
Group repeat events by shipment, location and risk type. Define recovery when the arrival is confirmed or stock is received.
Assign an owner, quiet hours and escalation. Preserve acknowledgement, snooze and action history.
A worked example
Three ETA changes for one shipment in an hour need not create three tasks. One incident can show the original promise, current date, affected SKUs and last check. A stopped carrier feed is a separate data-quality incident.
What to verify
A rule cannot confirm an event without its connected source. An alert prompts review; it does not authorize an automatic order or freight-price change.
Start with one rule, one owner and a clear closing condition in Yasnora AI.
