Packaging production line state monitoring and intelligent scheduling method based on internet of things

CN122243090APending Publication Date: 2026-06-19JIANGSU HONGYANG PACKAGING PRINTING CO LTD

Patent Information

Authority / Receiving Office: CN · China
Patent Type: Applications(China)
Current Assignee / Owner: JIANGSU HONGYANG PACKAGING PRINTING CO LTD
Filing Date: 2026-03-23
Publication Date: 2026-06-19

AI Technical Summary

Technical Problem

Existing technologies suffer from significant response delays when faced with disturbances such as emergency order insertions and sudden equipment failures. Centralized scheduling systems require excessive time to recalculate the global plan, and individual workstations lack autonomous collaboration capabilities, resulting in insufficient real-time response performance and robustness of the system.

Method used

A four-level collaborative architecture based on the Internet of Things (IoT) is constructed. By deploying IoT sensing terminals at workstations to collect data in real time, and using edge computing for localized processing, a perturbation-adaptive scheduling scheme is generated by combining a distributed negotiation protocol and a multi-agent reinforcement learning model, so as to achieve local autonomous decision-making and global collaboration.

Benefits of technology

It achieves second-level response capability under disturbance conditions, reduces scheduling delay, improves system robustness and flexibility, avoids resource conflicts and deadlocks, and improves production line operating efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure CN122243090A_ABST

Patent Text Reader

Abstract

This invention relates to the field of industrial automation and control system technology, and discloses a method for monitoring and intelligently scheduling the status of a packaging production line based on the Internet of Things (IoT). This method collects equipment, material, and task data in real time through IoT sensing terminals, and generates a dynamic state vector through edge computing. Each workstation autonomously generates local scheduling instructions based on this dynamic state vector and neighboring states through a distributed negotiation protocol. When a disturbance occurs, relevant workstations form temporary autonomous units, utilize a multi-agent reinforcement learning model to optimize task allocation and path planning online, generate a disturbance-adaptive scheduling scheme, and synchronize it to the central server to update the global plan. This method includes a sensing layer, an edge computing layer, a distributed scheduling layer, and a central coordination layer. This invention decentralizes scheduling decisions to the edge, achieving second-level disturbance response and improving the robustness and operational efficiency of the production line.

Need to check novelty before this filing date? Find Prior Art