Look-ahead Substrate Sequencing for Process Tool Idle Time Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In semiconductor manufacturing, process tools face inefficiencies due to complex and dynamic substrate handling sequences, leading to increased idle times and reduced throughput, particularly in cluster tools where static rules for substrate sequencing and buffering are inadequate for handling varying process times and product types.
Innovation Solution
Implementing an enhanced look-ahead functionality in the substrate handling system that predicts process times based on tool events from previously processed substrates, allowing for dynamic adaptation of transport activities and improved substrate sequencing to minimize idle times and enhance throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If static rules are used for substrate sequencing and buffering, then the control system remains simple and reliable, but idle times increase and throughput decreases
Solution Approach 1:
The patent implements dynamic substrate sequencing by replacing static rules with a predictive control system that adapts to varying process times. The system continuously monitors actual process times and dynamically adjusts the sequencing of substrates in the buffer, allowing the control mechanism to respond to real-time variations in process duration, thereby reducing idle times and improving throughput
Solution Approach 2:
The patent applies preliminary action by predicting future process times based on historical data and preparing the substrate sequence in advance. The system uses measured process times from previously processed substrates to forecast upcoming process durations, enabling the control mechanism to pre-arrange the optimal substrate sequence before idle times occur, thus minimizing waiting periods
2Adaptability or versatility
If process times vary due to different recipes and product types, then manufacturing flexibility improves, but predicting accurate start times for substrate handling becomes more difficult
Solution Approach 1:
The patent implements feedback by continuously measuring actual process times for different recipes and product types, then using this measured data to refine predictions for subsequent substrates. The system collects feedback from each processed substrate and adjusts its prediction model accordingly, maintaining high prediction accuracy despite varying manufacturing requirements and recipe complexities
Solution Approach 2:
The patent applies parameter changes by adapting the prediction model to account for different process parameters associated with various recipes and product types. The system modifies prediction parameters based on the specific recipe being used, substrate type, and process conditions, enabling accurate predictions across diverse manufacturing scenarios while maintaining flexibility
3Productivity
If more substrates are buffered to ensure continuous processing, then throughput may improve, but device complexity and resource requirements increase
Solution Approach 1:
The patent applies partial action by buffering only the necessary number of substrates required to bridge predicted idle times, rather than maintaining a large buffer for all possible scenarios. The system calculates the optimal buffer size based on predicted process time variations, maintaining sufficient substrates to prevent idle times while avoiding excessive buffering that would increase complexity and resource requirements
Data Source
AI summary
By providing a look-ahead functionality for a tool internal substrate handling system of process tools on the basis of a process history, the tool internal substrate sequencing may be significantly enhanced. The look-ahead functionality enables a prediction of process time of substrates currently being processed in a respective process module, thereby enabling the initiation of transport activity for substrate load operations in order to significantly reduce the overall idle time of process modules occurring during substrate exchange.