{"stable_id":"8e2cad283ed2ec2a","slug":"taiwan-tfda-border-food-risk-screening","company":"Taiwan Food and Drug Administration","workflow_name":"Risk-based screening for imported food inspection selection","function_code":"engineering","pattern_codes":["exception_based_operations","threshold_as_control","progressive_scope"],"changed_assumption":"Border inspectors do not have to rely primarily on random sampling for general-risk imports; historical violations, product attributes, importer data, and international food-safety alerts can target inspection capacity toward higher-risk batches.","evidence_strength":"verified","publication_tier":"showcase","freshness":"current","reviewed_at":"2026-09-06","updated_at":"2026-09-06","source_quality_summary":"2 peer reviewed, 2 primary, 1 independent; scoped inspection-yield outcomes are verified.","caveat_summary":"This is pre/post selected-population evidence, not a randomized causal estimate. The paper reports chi-square significance but notes pandemic, import-mix, regulatory, data-drift, and concept-drift factors that could affect year-to-year non-compliance rates. This is a TFDA-reported pre/post comparison for a specific fruit category, not a randomized causal study or an all-import outcome.","freshness_caveat":null,"workflow_summary":{"before":[{"actor":"Importer or inspection obligor","action":"Submitted the import inspection application, product information, import declaration copy, and required supporting.","actor_type":"control"},{"actor":"Import Food Inspection system","action":"Applied the regulatory inspection category and selected batches through random sampling in the risk-management step.","actor_type":"control"},{"actor":"Border inspector and laboratory","action":"Performed on-site checks and sampling for selected batches, then sent samples for laboratory analysis before import.","actor_type":"control"}],"hinge":"Scarce inspection capacity should follow predicted risk rather than random selection alone.","after":[{"actor":"Importer or inspection obligor","action":"Submits the same import application and documentation through the national import food information workflow.","actor_type":"system"},{"actor":"Import Food Inspection system","action":"Determines whether the product is subject to batch-by-batch inspection, regular random sampling, or enhanced random.","actor_type":"system"},{"actor":"Import Food Inspection system and border inspector","action":"Combine product risk items and inspector experience to determine inspection items, then send selected batches for.","actor_type":"system"},{"actor":"Import Food Inspection system","action":"Automatically falls back to random sampling for batches whose model computation takes more than one minute.","actor_type":"system"}],"decision_mode":"shared","decision_marker":"AI targets scarce inspection capacity; statutory authority stays human"},"before":[{"order":1,"actor":"Importer or inspection obligor","action":"Submitted the import inspection application, product information, import declaration copy, and required supporting documents online.","handoff_to":"Import Food Inspection system","control":"All batches still received document review under Taiwan's imported-food inspection rules."},{"order":2,"actor":"Import Food Inspection system","action":"Applied the regulatory inspection category and selected batches through random sampling in the risk-management step.","handoff_to":"Border inspector","control":"Regular random sampling generally covered 2% to 10% of products, while reinforced random sampling covered 20% to 50% and batch-by-batch inspection covered 100%."},{"order":3,"actor":"Border inspector and laboratory","action":"Performed on-site checks and sampling for selected batches, then sent samples for laboratory analysis before import approval.","handoff_to":"Import release or enforcement path","control":"Importation was permitted only after required inspections complied with regulations."}],"after":[{"order":1,"actor":"Importer or inspection obligor","action":"Submits the same import application and documentation through the national import food information workflow.","handoff_to":"Import Food Inspection system","control":"Document review and regulatory inspection categories remain the front-door control."},{"order":2,"actor":"Import Food Inspection system","action":"Determines whether the product is subject to batch-by-batch inspection, regular random sampling, or enhanced random sampling.","handoff_to":"Border Prediction Intelligent system","control":"BPI is applied to products that IFI places in general or enhanced random-sampling probability bands."},{"order":3,"actor":"Border Prediction Intelligent system","action":"Calculates real-time risk probability from domestic and international open data, food-cloud systems, product declaration data, violation history, and food-safety alerts; seven machine-learning algorithms vote on whether sampling should occur.","handoff_to":"Import Food Inspection system and border inspector","control":"The AI model assists and automates inspection-selection targeting; it does not replace the legal inspection regime."},{"order":4,"actor":"Import Food Inspection system and border inspector","action":"Combine product risk items and inspector experience to determine inspection items, then send selected batches for sampling and laboratory testing.","handoff_to":"Import release or enforcement path","control":"Human inspection judgment remains explicit for inspection items and exception handling."},{"order":5,"actor":"Import Food Inspection system","action":"Automatically falls back to random sampling for batches whose model computation takes more than one minute.","handoff_to":"Border inspector","control":"The fallback prevents model latency or failure from delaying border inspection."}],"decision_rights":"IFI retains the regulatory gate that classifies import applications into inspection procedures. BPI voting determines whether eligible general or enhanced random-sampling cases should be sampled. IFI and border inspectors retain judgment over inspection items, while statutory rules govern release, return, destruction, and escalation to higher sampling rates.","exception_path":"If a selected batch fails inspection, the importer must return or destroy the product unless TFDA approves corrective handling such as disinfection, reprocessing, safety measures, or labeling correction. The product's future sampling probability is increased, up to 100% batch-by-batch inspection. If BPI computation exceeds one minute, the system automatically uses random sampling so clearance is not delayed by model failure.","removed_work":["Using random sampling as the primary selector for eligible general-risk imported-food batches","Spending limited inspection capacity on lower-risk batches that the model can deprioritize","Relying only on historical inspection categories without real-time scoring from food-cloud and international alert data"],"outcomes":[{"metric":"S-type food non-compliance yield for EL-recommended inspected batches","baseline":"2019 pre-launch random sampling inspection unqualified rate was 2.09% for S-type food general sampling cases.","result":"EL-recommended inspected batches had unqualified rates of 5.10% in 2020, 6.36% in 2021, and 4.39% in 2022; each comparison to 2019 was reported as p < 0.001.","period":"2019 baseline compared with 2020-2022 post-launch model prediction sampling.","scale":"S-type food general sampling cases in Taiwan border inspection; 2019 table reports 29,573 inspection application batches, 3,157 overall sampled pieces, and 745 EL-suggested sampled pieces in 2020.","attribution_caveat":"This is pre/post selected-population evidence, not a randomized causal estimate. The paper reports chi-square significance but notes pandemic, import-mix, regulatory, data-drift, and concept-drift factors that could affect year-to-year non-compliance rates.","evidence_label":"verified"},{"metric":"Fresh/chilled/frozen fruit inspection effort, cost, and hit rate","baseline":"Before BPI, fresh/chilled/frozen fruit general random-sampling inspection averaged a 3.0% non-compliance hit rate.","result":"After BPI, the average inspection rate fell by 2.4 percentage points, inspection costs were reduced by more than NT$4 million, and the non-compliance hit rate rose from 3.0% to 3.8%, about a 30% improvement.","period":"MOHW/TFDA public release dated 2025-06-11, comparing pre- and post-BPI inspection results.","scale":"Fresh/chilled/frozen fruit under general random-sampling inspection, with more than 30,000 reported batches per year and 2% to 10% inspection probability.","attribution_caveat":"This is a TFDA-reported pre/post comparison for a specific fruit category, not a randomized causal study or an all-import outcome.","evidence_label":"verified"}],"executive_lesson":"Use AI to target scarce expert capacity, but keep statutory gates, human inspection judgment, and model-failure fallbacks explicit.","anti_pattern":"Treating a higher hit rate in model-selected inspections as proof of total-system causal impact without preserving the product scope, baseline period, and selected-population caveat.","questions_for_leaders":["Which inspection decisions should the model select directly, and which should remain human or statutory decisions?","What fallback is triggered when the model is slow, unavailable, or outside its validated product scope?","How often are risk factors refreshed to manage data drift and concept drift?","Which outcomes distinguish better targeting from a genuine reduction in downstream risk?"],"collections":["embodied-work","negative-results"],"bundle_version":"1.0.0","bundle_fingerprint":"sha256:8bbbaa076bde20fdad72e1c25c656436643f25efe84d187d0ec482f60acddc98","canonical_url":"https://brianletort.ai/transformations/taiwan-tfda-border-food-risk-screening"}