DAG-Based Smart Factory Scheduling Under Energy and Time Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing smart factory systems fail to consider intrinsic energy constraints and time limitations of manufacturing equipment (MEs), leading to energy depletion and production stagnation, while existing solutions neglect data sharing requirements and transmission power allocation, resulting in increased costs and delays.
Innovation Solution
An intelligent production method based on a Directed Acyclic Graph (DAG) is employed to determine and manage dependency relationships between MEs, construct a composite DAG, build device models, and optimize transmission power allocation using linear goal programming to minimize costs and ensure energy sufficiency and timely task completion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If data sharing between manufacturing equipment is increased to improve production efficiency, then productivity is improved, but energy consumption increases leading to energy depletion
Solution Approach 1:
The patent dynamically adjusts transmission power allocation ratios and task offloading decisions based on real-time energy status and task dependencies. By changing parameters such as power allocation ratios and offloading proportions, the system optimizes the balance between data sharing efficiency and energy consumption, preventing energy depletion while maintaining productivity.
Solution Approach 2:
The system implements dynamic task scheduling and power allocation that adapts to changing energy levels and production requirements. The DAG-based task scheduling dynamically adjusts execution sequences and resource allocation based on current system state, enabling the system to maintain optimal productivity across varying energy conditions.
2Speed
If transmission power is increased to accelerate data transmission between MEs, then speed is improved, but energy consumption increases
Solution Approach 1:
The patent optimizes transmission power allocation ratios as dynamic parameters based on task urgency, data size, and energy availability. The system adjusts power allocation to achieve minimum required transmission speeds while minimizing energy consumption, rather than using fixed high-power transmission.
Solution Approach 2:
The system applies partial power allocation rather than full power transmission by default. Transmission power is increased only partially and only when necessary based on task dependencies and deadlines, avoiding excessive energy consumption during low-priority data transmissions.
3Loss of time
If more tasks are executed locally on ME to reduce offloading overhead, then loss of time is reduced, but energy consumption increases
Solution Approach 1:
The system dynamically determines task offloading decisions based on real-time conditions including ME energy status, computational capabilities, task complexity, and deadline requirements. The offloading ratio is adjusted dynamically rather than using fixed rules, optimizing the trade-off between execution time and energy consumption for each task.
Solution Approach 2:
The patent introduces an intermediate optimization layer that mediates between local execution and cloud offloading. This layer evaluates task characteristics and system state to determine optimal offloading decisions, acting as an intermediary that balances the trade-off between local execution energy costs and offloading transmission costs.
4Productivity
If task scheduling does not consider energy constraints, then productivity is improved, but reliability deteriorates due to energy depletion and production stagnation
Solution Approach 1:
The patent performs preliminary energy assessment and task scheduling that proactively prevents energy depletion before it occurs. The system evaluates energy constraints in advance when creating production schedules, ensuring that energy requirements are met before committing to task execution sequences, thus preventing production interruptions.
Solution Approach 2:
The system builds in energy buffers and contingency plans in the task scheduling process. By预留 (reserving) energy margins and planning for energy constraints beforehand, the system creates a cushion that prevents production failures even when energy consumption exceeds initial estimates or unexpected tasks arise.
Data Source
AI summary
In an intelligent production method, external dependency relationships between different ME and internal dependency relationships within a same ME are determined. A composite DAG, a device model, a smart factory model are constructed. Then, the smart factory model is solved using a linear goal programming algorithm with an objective of minimizing a total cost, to obtain an optimal transmission power allocation ratio and an optimal offloading decision. An optimal production plan is determined based on the optimal transmission power allocation ratio and the optimal offloading decision. And an intelligent production is executed according to the optimal production plan.