Neural Network KPI Control for Bottleneck WIP Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methodologies are inadequate for identifying key factors contributing to cycle time reduction in complex manufacturing processes, such as semiconductor wafer fabrication, where dozens of stages with multiple steps are involved, leading to unreliable results.
Innovation Solution
A system and method that groups tools by similar process steps, collects and analyzes process profile data to calculate key performance indicators (KPIs), uses a neural network model to assess the impact of KPIs on work-in-progress (WIP), and adjusts tool group operations to reduce total WIP, focusing on standard deviation of bottleneck tool groups to improve cycle time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If traditional methodologies (linear programming, computer simulation, queue model) are used to analyze complex manufacturing processes, then the analysis can be performed with simple tools, but the reliability of identifying key cycle time factors deteriorates due to process complexity
Solution Approach 1:
The patent replaces traditional mechanical/mathematical analysis methods (linear programming, queue models, simulation) with an AI-based neural network system. The neural network learns patterns from historical manufacturing data to identify key cycle time factors, substituting conventional analytical mechanics with intelligent computational mechanics that can handle complex, non-linear relationships in semiconductor manufacturing processes.
2Adaptability or versatility
If the manufacturing process includes dozens of stages with multiple steps (as in semiconductor wafer fabrication), then the manufacturing capability and product complexity increase, but the difficulty of detecting and measuring key cycle time factors increases
Solution Approach 1:
The patent introduces an intermediary layer between the complex manufacturing process and the analysis system. The neural network acts as an intermediary that processes raw manufacturing data, extracts relevant features, and identifies key cycle time factors. This intermediary transforms the complex, high-dimensional manufacturing process data into manageable insights about critical process steps and bottlenecks.
Solution Approach 2:
The patent creates a virtual copy or digital twin of the manufacturing process through data collection and neural network modeling. Instead of directly analyzing the physical complex process, the system creates a computational model that replicates process behavior, allowing analysis of cycle time factors in the virtual model without disrupting the actual manufacturing process.
3Adaptability or versatility
If more process stages and steps are added to handle complex manufacturing requirements, then the manufacturing versatility improves, but the cycle time increases due to more work-in-process
Solution Approach 1:
The patent implements a feedback mechanism where the neural network continuously monitors manufacturing process data, identifies bottlenecks and cycle time drivers, and provides insights for process optimization. This feedback loop enables dynamic adjustment of process parameters and resource allocation to minimize cycle time while maintaining the necessary manufacturing versatility for complex products.
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.