When an aircraft is grounded, every hour matters.
But it is too simplistic to say that customs paperwork is always the reason an aircraft stays on the ground. The aviation industry is still dealing with wider supply-chain volatility, aircraft and parts delivery delays, engine problems, and spare-parts shortages. What is true — and operationally important — is narrower: once a part is available and has to cross a border, customs documentation can become a critical path item.
The document bottleneck no one talks about
AOG logistics teams are built for urgency. The industry has invested for years in AOG desks, round-the-clock supplier coverage, and rapid parts sourcing because aircraft-on-ground events do not tolerate slow handoffs.
The friction often starts when the shipment becomes a cross-border move. At that point, the work is no longer only about finding the part; it is about assembling the right document set, validating the data, and getting the filing right for the jurisdictions involved. Depending on the route, commodity, and customs treatment, that package may include:
- A commercial invoice and supporting shipment data.
- An air waybill or master air waybill for air freight.
- An EASA Form 1 or FAA Form 8130-3, where applicable to the part.
- Proof of origin, where tariff treatment or the importing jurisdiction requires it.
- Entry Summary Declaration data for EU-bound or EU-transiting cargo under ICS2.
- A Shipper's Declaration for Dangerous Goods when the consignment contains dangerous goods.
That is the real bottleneck: not the existence of a single PDF, but the need to reconcile multiple files, from multiple parties, into a broker-ready and regime-correct customs package.
Why generic customs AI does not fully solve it
The first generation of document AI mostly focused on extraction: take a file in, return structured fields out. That is useful, but AOG customs work is more operational than that.
The hard part is usually a combination of four jobs:
- Receiving the right documents from different parties and channels.
- Correlating the shipment data across those documents.
- Validating the file against the actual regime in play.
- Producing output that the broker, declarant, or customs workflow can use immediately.
That third step matters more than many teams expect. In Canada, commercial duties and taxes are administered through CARM. In the EU, ICS2 requires safety and security data through the Entry Summary Declaration, including pre-loading and pre-arrival reporting for air cargo. In the UK, customs declarations now run through CDS. And CBAM is not a blanket customs rule for all imports: from 1 January 2026 it applies to defined in-scope goods, initially including cement, iron and steel, aluminium, fertilisers, electricity, and hydrogen, with specific obligations for importers above the 50-tonne threshold.
So the real gap is not just "can the software read the document?" It is "can the workflow assemble, validate, and route the shipment correctly for the jurisdiction that actually applies?"
The workflow-first approach
For most AOG teams, the best solution is rarely "another portal."
The better operating model is workflow-first: keep document intake inside the tools the team already uses, automate the extraction and cross-checking in the background, and return a broker-ready package without forcing people into a new process at the worst possible moment.
That does not mean the interface matters more than the compliance logic. It means the automation has to fit the operating reality of urgent logistics work: fragmented document arrival, shifting shipment details, and very little tolerance for rework.
What better execution looks like in practice
A stronger AOG customs workflow usually looks like this:
| Step | Manual workflow | Workflow-first automation |
|---|---|---|
| Document intake | Teams chase files across inboxes, threads, and forwarders | Documents are captured as they arrive |
| Data extraction | Staff manually re-key invoice, AWB, and release-certificate data | Core fields are extracted automatically |
| Cross-checking | Teams compare part, shipper, consignee, route, and origin details by hand | Exceptions are flagged for review |
| Regime handling | Staff translate the shipment into broker-ready data for the relevant filing path | Output is structured for the jurisdiction in play |
| Final review | Human review happens late, after manual prep | Human review focuses on exceptions and release readiness |
The point is not to remove people from the process. The point is to remove avoidable document handling so skilled operations staff spend their time on judgment, escalation, and release readiness instead of repetitive data entry.
The compliance layer cannot be optional
Speed matters. But speed without compliance is not an AOG advantage.
If the file is wrong, the shipment can still be delayed, corrected, or exposed to penalties. In Canada, proof of origin must be furnished for imported goods, and penalties can apply when required proof is not presented on request. Under CBAM, incorrect or missing reporting can also trigger penalties.
That is why the right architecture does not treat compliance as a second pass. It validates the file as part of the workflow itself:
- Is the origin evidence actually appropriate for the tariff treatment claimed?
- Does the shipment require ICS2 ENS data because it is entering or transiting the EU?
- Is the declaration path in the UK aligned with CDS?
- Is CBAM even relevant to this shipment, or is it out of scope?
- Does the consignment include dangerous goods that require a DGD?
In AOG, the fastest process is the one that does not have to be redone.
What this means for your ops team
If your team is still spending its time collecting files, copying fields between documents, and translating shipment data manually for brokers and customs workflows, there is a meaningful opportunity to reduce delay risk.
The practical benchmark is not whether an AI tool can read a PDF. It is whether it can help your team move from fragmented documents to a compliant, broker-ready customs package with less rework and fewer avoidable handoffs.
That is the standard that matters in AOG: faster execution, yes — but only if the workflow remains accurate enough to clear.
Doana is an AI customs agent built specifically for aviation, freight, and supply chain teams. It connects to Outlook, Gmail, Slack, Teams, and WhatsApp and processes customs documents — including EASA Form 1, commercial invoices, and AWBs — with built-in CARM, ICS2, CBAM, and UK CDS compliance checks.