Neural Network KPI Control for Semiconductor WIP Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methodologies, such as linear programming and queue models, are inadequate for identifying key factors that reduce cycle time in complex manufacturing processes like semiconductor manufacturing, where hundreds of steps are involved, making it difficult to optimize operations effectively.
Innovation Solution
A system and method that groups tools by similar process steps, uses managing computers to collect and analyze process profile data, calculates key performance indicators (KPIs) like standard deviation of output from bottleneck tool groups, and employs a neural network model to analyze the impact of KPIs on work-in-progress (WIP), allowing for data-driven adjustments to reduce total WIP and improve cycle time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional methodologies (linear programming, queue models) are used to analyze manufacturing processes, then the analysis approach is simple and familiar, but the reliability of identifying key cycle time factors is insufficient for complex processes with hundreds of steps
Solution Approach 1:
The patent replaces traditional mechanical/mathematical analysis methods (linear programming, queue models) with a neural network-based intelligent system. The neural network automatically learns patterns and relationships in manufacturing process data, providing reliable identification of key cycle time factors without requiring complex manual modeling of hundreds of process steps.
Solution Approach 2:
The patent creates a virtual model (neural network) that copies and simulates the complex manufacturing process relationships. Instead of directly analyzing the physical process with traditional methods, the neural network learns from historical data and creates a digital representation that can reliably predict and identify key factors affecting cycle time.
2Adaptability or versatility
If the number of process steps is increased to hundreds of steps to cover all manufacturing operations, then the completeness of process coverage is improved, but the complexity of the manufacturing process increases making traditional methodologies unreliable
Solution Approach 1:
The patent segments the complex manufacturing process into manageable components by having the neural network automatically identify and focus on key process steps and parameters that most impact cycle time. Instead of treating all hundreds of steps equally, the system segments attention to the critical few factors that drive performance.
Solution Approach 2:
The patent transforms the approach by changing from fixed traditional analysis parameters to dynamic parameters learned by the neural network. The system automatically adjusts which process parameters are most relevant based on the data, allowing complete process coverage while managing complexity through intelligent parameter selection.
3Measurement precision
If more process data and KPIs are collected and analyzed, then the accuracy of identifying key factors is improved, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent applies partial action by having the neural network focus computational resources on analyzing the most impactful KPIs and process parameters. Rather than exhaustively processing all possible data with equal depth, the system identifies and deeply analyzes the critical subset of data that most influences cycle time, achieving high accuracy with optimized computational power.
Data Source
AI summary
A method for improving a cycle time of a process of a product is provided. The method includes: collecting process profile data from a plurality of tool groups running the process, and calculating values of a plurality of key-performance-indicators (KPIs) of each tool group including calculating a standard deviation of an output of a stage of a bottleneck tool group of the tool groups; feeding the values of the KPIs and a work-in-progress (WIP) of each tool group into a neural network model in order to output an impact on the WIP for each KPI of each tool group by the neural network model; selecting a set of major KPIs of each tool group from the KPIs according to the impact of each tool group; and controlling the tool groups according to the impact of the set of major KPIs of each tool group in order to reduce a total WIP.