AI-Assisted Passenger Goods Inspection at Ngurah Rai International Airport - A Simplified Design Science Research Study
Year: 2025
Type: article
Abstract: Background: Ngurah Rai International Airport handles over 6 million international passengers annually, creating significant pressure on customs inspection operations. Traditional manual inspection processes struggle with increasing passenger volumes, inconsistent decision-making, and inadequate documentation systems. Objective: This Design Science Research study implements and evaluates an AI-assisted customs inspection Telegram bot at Ngurah Rai Customs Office to improve passenger goods examination efficiency, accuracy, and documentation quality through conversational AI interface. Method: An 8-week field study using Simplified Three-Cycle DSR methodology with 25 customs officers. The study progressed through Relevance Cycle (problem identification), Design Cycle (iterative AI solution development), and Rigor Cycle (full deployment and evaluation) using mixed-methods data collection including performance surveys and qualitative interviews. Results: AI implementation resulted in 47% reduction in inspection time (from 4.2 to 2.2 minutes average), 52% decrease in classification errors, and 28% improvement in risk detection accuracy. Officer satisfaction increased significantly with 89% expressing intention to continue using the system post-study. Conclusions: The AI-powered Telegram bot successfully enhanced airport customs operations without disrupting established workflows. Conversational AI interfaces proved highly effective for routine passenger inspections in high-pressure government environments, providing a scalable model for Indonesian customs modernization.
Authors: Setiyadi, Kurniawan, Hasbullah, Sasonoto
Institution: Ministry of Finance
Language: English
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W6968036264
FWCI:
Cites:
Cited by:
Related to: