Semiconductor Module Scheduling Using Rule-Based Task Execution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semiconductor process apparatuses face challenges in efficiently scheduling multiple materials due to complex algorithms, manual setting of execution times, and increased idle time for modules, which affects production capacity.
Innovation Solution
A scheduling control method for a semiconductor process apparatus that acquires module information one module at a time, determines task executability based on preset rules and real-time module and material information, and controls modules to execute tasks accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If complex scheduling algorithms (search tree or Petri network) are used to calculate transfer paths for all materials, then scheduling completeness is improved, but calculation time increases significantly and device complexity increases
Solution Approach 1:
The patent divides the scheduling problem into independent module-level decisions. Instead of calculating transfer paths for all materials globally using complex algorithms, the system evaluates each module independently to determine if its specified task is executable based on local state information and preset rules. This segmentation reduces calculation complexity and time while maintaining scheduling effectiveness.
Solution Approach 2:
Each module performs self-evaluation of task executability based on its own state information and preset rules, without requiring centralized complex algorithmic coordination. The module autonomously determines whether to execute its specified task by checking local conditions, reducing the need for system-wide calculation while ensuring proper scheduling decisions.
2Measurement precision
If manual setting of execution time for each module is performed, then scheduling control precision is improved, but manpower consumption increases and error probability increases
Solution Approach 1:
The system eliminates manual setting of execution times by using preset rules that automatically determine task executability based on real-time module state information. The modules self-regulate their execution decisions through programmed logic rather than manual configuration, reducing manpower consumption and setup effort while maintaining control precision through automated state-based decision-making.
Solution Approach 2:
The patent transitions from fixed manually-set execution times to dynamic parameter evaluation based on real-time module state. Instead of relying on predetermined execution times set manually, the system uses preset rules that evaluate current module conditions (idle state, material existence, task completion status) to dynamically determine task executability, reducing setup effort while maintaining precision through adaptive decision-making.
3Measurement precision
If transfer paths are calculated for all materials before execution, then scheduling accuracy is improved, but module idle time increases and productivity decreases
Solution Approach 1:
The system performs preliminary evaluation of task executability using preset rules and current module state information before execution, but does not require complete pre-calculation of all transfer paths. Modules can begin execution once their specific task conditions are met, rather than waiting for global path calculation completion, reducing idle time while maintaining scheduling accuracy through rule-based decision-making.
Solution Approach 2:
The patent implements dynamic scheduling where module task execution decisions are made based on real-time state information and preset rules, rather than static pre-calculated transfer paths for all materials. This allows modules to dynamically adjust their execution timing based on current conditions, reducing idle time and improving productivity while maintaining scheduling accuracy through continuous state evaluation.
Data Source
AI summary
A scheduling control method for a semiconductor process apparatus and a semiconductor process apparatus are provided. The semiconductor process apparatus includes a plurality of modules that are configured with corresponding specified tasks. The method includes: for the plurality of modules, acquiring module information corresponding to the modules one module by one module, and determining whether the specified tasks corresponding to the modules are executable based on preset rules and the module information corresponding to the modules, with the module information including module state information and material existence information and the preset rules including rules for determining whether the modules execute the corresponding specified tasks. If it is determined that the specified tasks corresponding to the modules are executable, the method includes controlling the modules to execute the corresponding specified tasks.