evidoria

← Back to browse

Good practice Imported

Tianxuan — China Customs' Nationwide AI Targeting Model, Documented by the World Customs Organization

China · Beijing · See the China profile · See the Beijing profile

Evidence: Observational / pre–post Top 22% 73/100 · Ask Evidence Copilot about this practice

China's customs authority (GACC) has deployed its 'Tianxuan' AI targeting model at 269 sites nationwide. A 2025 WCO case study reports it drove 4.32% of all targeting operations in the first 11 months of 2024, with a seizure rate 7 percentage points higher than manual targeting.

269
Operational sites (November 2023)
4.32%
Share of targeting operations via Tianxuan (first 11 months of 2024)
+7 percentage points
Seizure-rate improvement vs manual targeting
Tianxuan — China Customs' Nationwide AI Targeting Model, Documented by the World Customs Organization

Details

Maturity
Established
Promoter
General Administration of Customs of China (GACC)
Period
2023–2024
Keywords
customs, trade facilitation, risk management, border control

Context

Since a 2017 pilot at Huangpu Customs, GACC has scaled its 'Tianxuan' machine-learning targeting model nationwide to 269 operational sites by November 2023, as part of its 'smart customs' modernisation programme.

Results

A 2025 WCO case study reports Tianxuan accounted for 4.32% of all customs targeting operations in the first 11 months of 2024, with a seizure rate 7 percentage points higher than manual targeting.

Conclusions

The WCO published the material as a case study for other member customs administrations, but neither GACC nor the WCO discloses the model's scoring criteria or an independent appeals/oversight mechanism.

Implementation
Indicative cost
High (€500k–€5M)
Time to results
Long (> 3 years)
Staffing & skills
General Administration of Customs of China (GACC) targeting analysts

Conditions for success

  • nationwide rollout across operational sites
  • sustained machine-learning model retraining on declarations/trade data

Where it fits

Governance type
national
Scale
national

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful