TIME-AWARE ALLOCATION OF CUSTOMS INSPECTION RESOURCES ACROSS RISK LANES: DECLARATION-LEVEL EVIDENCE FROM UZBEKISTAN

This article examines how customs authorities can shorten clearance times without reducing the detection of violations by redistributing fixed inspection capacity across risk lanes. The empirical basis is a declaration-level dataset from the Customs Committee of Uzbekistan, comprising 8,000 declarations lodged at 12 customs posts during 2023–2026. A log-logistic accelerated failure time model shows that assignment to the red lane multiplies expected clearance time by 2.97, while pre-arrival lodging reduces it by 29.7%. A linear programming model is proposed to rank declarations by the number of detected violations per hour of additional clearance time. With the same 1,820 inspections and 786 detected violations, the model reduces the 90th percentile of clearance time by 21.3 minutes and by 64.2 minutes when combined with an increased share of pre-arrival declarations.