Detect and resolve air freight disruptions faster.

Air freight disruption management is not a status colour on a dashboard. It is a sequence: detect the event, identify the shipment, assess operational impact, recommend recovery, take an operator decision, then communicate. The interactive air freight demo walks that sequence on a simulated cargo roll.

UpliftOS is designed around the events that actually stop air cargo moving:

Flight cancellation already creates a multi-shipment disruption, recovery plan and inbox work in the product. Cargo roll from an airline email is the primary single-file demo: classify the message, match the AWB, mark the shipment disrupted, recommend the next flight, and wait for operator approval.

Detection can start from an operational email, a flight status change, a cutoff that has passed, or a document that is still invalid. Assessment is shipment-specific: required delivery, booked routing, remaining capacity on recovery flights, and whether dangerous goods or SLA constraints change the options.

Recommendation is not a paragraph of advice. It is a concrete next booking or hold, with impact in hours, presented in the Operations Inbox. Recovery executes only after an operator approves. Communication is drafted from the new state; it is not auto-sent.

A roll is the textbook case: the airline keeps the booking reference and moves the freight to a later flight, often the same service tomorrow. The operator still has to find every HAWB under the MAWB, revise ETA, check the consignee SLA, and send a message that does not contradict the new routing. UpliftOS matches the AWB, opens a critical inbox item, and proposes the recovery flight for approval.

A delay on QF81 is not one shipment. It is every file booked on that uplift, plus any onward connection that will break. Cancellation is worse: capacity disappears and recovery has to be planned across the affected set. UpliftOS groups those shipments, calculates impact and proposes recovery rather than leaving each operator to rebuild the same spreadsheet.

Rejected nominations and cargo that misses lodge close produce the same customer outcome as a roll — the freight does not fly as promised — but the operational cause is different. The software should record the cause, not only the delay, so the branch can see whether the leak is airline capacity, warehouse timing or documentation.

Missing commercial invoices, packing lists or DG declarations stop lodgement. Unconfirmed collections stop the file even earlier. These are disruptions in operational terms even when the aircraft is on time. UpliftOS raises them in the same inbox as airline events so they compete for attention on severity and cutoff, not on which mailbox they arrived in.

AI automation prepares the work. Humans approve recovery. That is the current product policy, documented as observe → reason → recommend → approve → execute. Later customers may configure constrained autonomy; it is not enabled by default.