South KoreaLogistics and transport2025 – ongoing

CJ Logistics

AI dispatch, load planning and robot automation

CJ Logistics is moving phone-based truck dispatch to forecasting and optimisation; a load-planning agent cut calculation time from an hour to ten minutes

In progressCompany's own claim

Problem

Trunk-line trucks between about 140 centres were dispatched by phone, relying on dispatchers' experience. Planning how to load vehicles and pallets also took time and left unused capacity.

Approach

The company's logistics technology institute (TES) is replacing phone dispatch with a demand-forecasting model and an optimisation engine fed by scan, master and event data. For contract-logistics customers it built an AI agent that calculates loading plans, and it is testing language-model agents that write optimisation models. At its Incheon global distribution centre it put a mobile, vision-based robot palletizer into service, scaled step by step from one to four pallets at a time.

Results

1 h → 10 mintime to calculate a loading plan with the AI agent for contract-logistics customersCompany's own claim[1]

~140centres whose trunk-line dispatch was coordinated by phone before the changeCompany's own claim[1]

4pallets the mobile robot palletizer handles at the same time at the Incheon centreCompany's own claim[2]

Takeaway

CJ Logistics developed a load-planning agent while moving dispatch from phone calls to forecasting and optimisation. The company's figures cover individual tools; it has not published results for the dispatch system as a whole.

Sources

CJ대한통운이 AI로 물류를 최적화하는 법
바이라인네트워크28 Nov 2025News article
CJ대한통운, 현장 맞춤형 인공지능 로봇 자동화로 물류 생산성 향상
토요경제17 Jul 2025News article

Source summaries

[1]바이라인네트워크28 Nov 2025

A CJ Logistics optimisation researcher spoke at Byline Network's retail and logistics tech conference in November 2025. He described moving trunk-line dispatch across about 140 centres from phone calls to forecasting and optimisation. He said a load-planning agent cut calculation time from about an hour to under ten minutes and improved leftover volume by about 13%. In a trial, language-model agents shortened optimisation model development from months to one or two weeks.

[2]토요경제17 Jul 2025

Reports that CJ Logistics put a mobile, AI vision-based robot palletizer into use at its Incheon global distribution centre, scaling from one to four pallets at a time after checks at each stage. The company says productivity improved and worker fatigue fell, without giving figures.

Full details are in the original sources.

Added 25 Oct 2026 · Updated 25 Oct 2026