Proceedings of the 2026 7th International Conference on Management Science and Engineering Management (ICMSEM 2026)

2026 7th International Conference on Management Science and Engineering Management (ICMSEM 2026)

📍Shenyang, China🗓️ 22-24 May 2026

AI-Driven Construction of an Intelligent Customs System in the Guangdong-Hong Kong-Macao Logistics Park: Innovations in Bonded Ledger Management and Cross-Border Data Integration

Authors
Shuangtao Wei1, Yuefeng Zou1, Chengguo Han1, *, Xun Su1
1Guangdong University of Science and Technology, Dongguan, Guangdong, 523079, China
*Corresponding author. Email: 1819777573@qq.com
Corresponding Author
Chengguo Han
Available Online 8 September 2026.
DOI
10.2991/978-94-6239-758-3_9How to use a DOI?
Keywords
Intelligent Customs System; Artificial Intelligence (AI); Guangdong-Hong Kong-Macao Logistics Park; Microservices Architecture; Cross-border Data Integration; Bonded Ledger Management
Abstract

Targeting the “three complexities” pain points of the Guangdong-Hong Kong-Macao Logistics Park—diversified cross-border logistics supervision, varied business formats, and fragmented systems—and in response to the mandatory requirements of the General Administration of Customs Announcement No. 174 of 2024 on dynamic balancing of bonded account books and fully electronic management, this paper constructs an AI-driven Intelligent Customs System (ICS). The system deeply integrates OCR, natural language processing, Transformer encoding mapping, and a dynamic account book verification engine. Based on a cloud-native microservice architecture, it realizes automation of cross-border declarations and data synchronization between the Warehouse Management System (WMS) and customs account books, ultimately achieving secure dual-declaration interaction. This study selects a dataset consisting of 100,000 trade records and 50,000 release verification documents from 52 enterprises in the park between January 2024 and March 2025, with the dataset split into 80% training set, 10% validation set, and 10% test set. Traditional manual verification and general OCR are used as baseline benchmarks. Practical verification shows that the system reduces the account book discrepancy rate from 18% to 0.3%, achieves a 95% accuracy rate in HS code conversion, cuts document rejection rate by 82%, improves release verification document processing efficiency by 50%, and lowers labor costs by 60%. These results significantly enhance cross-border logistics processing efficiency and customs supervision compliance, providing a replicable practical paradigm for the construction of smart customs in the Guangdong-Hong Kong-Macao Greater Bay Area.

Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

Volume Title
Proceedings of the 2026 7th International Conference on Management Science and Engineering Management (ICMSEM 2026)
Series
Advances in Economics, Business and Management Research
Publication Date
8 September 2026
ISBN
978-94-6239-758-3
ISSN
2352-5428
DOI
10.2991/978-94-6239-758-3_9How to use a DOI?
Copyright
© 2026 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

TY  - CONF
AU  - Shuangtao Wei
AU  - Yuefeng Zou
AU  - Chengguo Han
AU  - Xun Su
PY  - 2026
DA  - 2026/09/08
TI  - AI-Driven Construction of an Intelligent Customs System in the Guangdong-Hong Kong-Macao Logistics Park: Innovations in Bonded Ledger Management and Cross-Border Data Integration
BT  - Proceedings of the 2026 7th International Conference on Management Science and Engineering Management (ICMSEM 2026)
PB  - Atlantis Press
SP  - 85
EP  - 94
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-758-3_9
DO  - 10.2991/978-94-6239-758-3_9
ID  - Wei2026
ER  -