Adaptive Scheduling Frame for Dynamic Custom Production Events
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional unified computing frames for personalized customized production lack intelligence and adaptability to the high discreteness and complexity of production demands, as well as the high frequency of dynamic events, leading to inefficient scheduling and high labor costs.
Innovation Solution
An adaptive-learning intelligent scheduling unified computing frame and system that employs deep reinforcement learning and neural networks to classify dynamic events, optimize scheduling plans, and automate equipment deployment, thereby improving optimization efficiency and reducing labor costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional unified scheduling optimization method is adopted, then scheduling plan can be generated, but it cannot adapt to high discreteness and complexity of personalized customized production demands
Solution Approach 1:
The patent segments the scheduling system into multiple specialized modules including static scheduling module, dynamic event classification module, and targeted optimization modules. Each module handles specific aspects of the scheduling problem, allowing the system to adapt to complex personalized production demands through coordinated operation of specialized components rather than a monolithic approach
Solution Approach 2:
The patent implements dynamic adaptability through real-time event classification and dynamic optimization. The system continuously monitors production events, classifies them using deep reinforcement learning, and adjusts scheduling strategies dynamically. This allows the system to adapt to changing production demands while maintaining manageable complexity through automated decision-making
2Extent of automation
If manual scheduling plan creation is used, then scheduling decisions can be made, but labor costs are high and intelligence level is low
Solution Approach 1:
The patent implements self-service through automated scheduling decision-making. The system uses deep reinforcement learning models to automatically classify dynamic events and generate optimization strategies without human intervention. The intelligent agent continuously learns from production data and makes autonomous scheduling decisions, eliminating manual labor while reducing decision time through automated real-time processing
Solution Approach 2:
The patent incorporates feedback mechanisms where the system continuously monitors production status, compares actual outcomes with predicted outcomes, and uses deep reinforcement learning to update optimization strategies. This closed-loop feedback enables the system to learn from experience and improve scheduling decisions over time, increasing intelligence while maintaining rapid automated response
3Reliability
If existing computing frame is used, then basic production functions are supported, but coordination between equipment and modules is insufficient
Solution Approach 1:
The patent creates a universal computing frame that coordinates multiple equipment types and modules through standardized interfaces and unified optimization strategies. The system uses deep reinforcement learning to generate coordination strategies that work across different equipment and module configurations, improving reliability through consistent coordination while managing complexity through standardized multi-functional protocols
Solution Approach 2:
The patent introduces an intelligent scheduling system as an intermediary between equipment and modules. This intermediary uses deep reinforcement learning to process information from various sources, make coordination decisions, and issue control commands. The intermediary manages complexity by centralizing decision-making logic while improving reliability through intelligent coordination of all equipment and modules
Data Source
AI summary
An adaptive-learning intelligent scheduling unified computing frame and system for industrial personalized customized production that are based on a deep neural network and reinforcement learning. An optimization algorithm is selected by automatic decision-making for a global customized production task with an industrial big data module at the bottom as an information basis, and a global optimal static scheduling plane is generated; a current dynamic event is monitored in real time; if not monitored, the global optimal static plan is executed sequentially; when a dynamic event impact requiring dynamic scheduling optimization is monitored, information of the current dynamic event is interpreted and classified, and corresponding optimization algorithms are automatically selected for dynamic scheduling optimization; and a dynamic scheduling scheme is evaluated by a subsequent module, an optimization scheme is regenerated or a most suitable optimization algorithm is automatically decided, and an equipment deployment sequence is generated for an automatic deployment.