An agent and large language model-based material distribution scheduling method, system and intelligent terminal

By integrating intelligent agents with large language models to produce end-to-end data, and performing task objective orchestration, reasoning, and flexible scheduling rearrangement, the problem of insufficient data integration in traditional material distribution scheduling is solved, thereby improving material distribution efficiency and resource utilization.

CN122243154APending Publication Date: 2026-06-19ZHEJIANG CHINAJEY SOFTWARE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG CHINAJEY SOFTWARE TECH CO LTD
Filing Date
2026-05-25
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Traditional material distribution scheduling methods cannot effectively integrate a unified data foundation and real-time sensing data across the entire production chain. This results in a lack of comprehensive and real-time data for the arrangement of distribution task objectives, an inability to accurately match available distribution resources with reachable paths, and a lack of flexible scheduling and rescheduling capabilities, leading to low material distribution efficiency and low resource utilization.

Method used

By using an agent-based and large language model approach, the unified data foundation of the entire production chain and real-time perception data are integrated into global data. This enables the orchestration and reasoning of task objectives and flexible scheduling rearrangement. Combined with feedback data from the collaborative end, the scheduling weights are optimized to generate the final scheduling instructions.

Benefits of technology

It ensured the rationality and timeliness of material distribution tasks, improved execution efficiency and resource utilization, and guaranteed the efficient and coordinated operation of the entire production chain.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122243154A_ABST
    Figure CN122243154A_ABST
Patent Text Reader

Abstract

This invention relates to the field of resource scheduling technology, and proposes a material delivery scheduling method, system, and intelligent terminal based on intelligent agents and large language models. The method includes: fusing a unified data foundation across the entire production chain with real-time sensing data into global data; in the intelligent agent, based on the global data, performing orchestration and reasoning on task objectives to obtain candidate delivery sequences for the task objectives; performing flexible scheduling rearrangement on the candidate delivery sequences and encoding the rearrangement results into scheduling instructions; optimizing the weights of the flexible scheduling rearrangement based on feedback data from the collaborating end; and generating the final scheduling instructions for the task objectives based on the optimized weights. This invention can improve the efficiency of material delivery scheduling.
Need to check novelty before this filing date? Find Prior Art