Data-Driven Scheduling of Semiconductor Manufacturing Systems

Preface
Contents
1 Scheduling of Semiconductor Manufacturing System
1.1 Semiconductor Manufacturing Process
1.2 Scheduling of Semiconductor Manufacturing System
1.2.1 Scheduling Characteristics
1.2.2 Scheduling Types
1.2.3 Scheduling Methods
1.2.4 Evaluation Indicators
1.3 Scheduling Development Trend of Semiconductor Manufacturing System
1.3.1 Data Preprocessing of Complex Manufacturing System
1.3.2 Data-Based Scheduling Modeling
1.3.3 Data-Based Scheduling Optimization
1.3.4 Analysis of Research Status
1.4 Summary
References
2 Data-Driven Scheduling Framework of Semiconductor Manufacturing System
2.1 Design of Data-Driven Scheduling Framework
2.2 Data-Based Scheduling Architecture of Complex Manufacturing System
2.2.1 Overview of DSACMS
2.2.2 Formal Description of DSCAMS
2.2.3 DSACMS-Based Modeling and Optimization of Scheduling for Complex Manufacturing Systems
2.2.4 Key Technologies in DSACMS
2.3 Application Examples
2.3.1 Overview of Fabsys
2.3.2 Object-Oriented Simulation Model of Fabsys (OOSMfab)
2.3.3 Data-Driven Forecasting Model in FabSys
2.4 Summary
References
3 Data Preprocessing of Semiconductor Manufacturing System
3.1 Introduction
3.2 Data Standardization
3.2.1 Data Normalization Rules
3.2.2 Correction of Abnormal Values for Variables
3.3 Filling of Missing Data
3.3.1 Filling Method for Missing Data
3.3.2 Memetic Algorithm and Memetic Calculation
3.3.3 Attribute Weighted K Nearest Neighbor Missing Value Filling Method (KNN) Based on Gaussian Mutation and Depth First Search (GD-MPSO): GD-MPSO-KNN
3.3.4 Numerical Verification
3.4 Outlier Detection Based on Data Clustering Analysis
3.4.1 Outlier Detection Based on Data Clustering
3.4.2 K-Means Clustering
3.4.3 Data Clustering Algorithm Based on GS-MPSO and K-Means Clustering (GS-MPSO-KMEANS)
3.4.4 Numerical Verification
3.5 Redundant Variable Detection Based on Variable Clustering
3.5.1 Principal Component Analysis
3.5.2 Variable Clustering Based on K-Means Clustering and PCA
3.5.3 Variable Clustering Algorithm Based on MCLPSO (MCLPSO-KMEANSVAR)
3.5.4 Numerical Verification
3.6 Summary
References
4 Correlation Analysis of Performance Index of Semiconductor Production Line
4.1 Long-Term and Short-Term Performance Indicators of Semiconductor Manufacturing System
4.2 Statistical Analysis of Performance Indicators for Semiconductor Production Lines
4.2.1 Short-Term Performance Indicators
4.2.2 Long-Term Performance Indicators
4.3 Correlation Analysis of Performance Indicators Based on Correlation Coefficient Methods
4.3.1 Block Diagram of Correlation Analysis
4.3.2 Correlation Analysis of Performance Indicators Considering Working Conditions
4.3.3 Correlation Analysis of Performance Indicators Considering Dispatching Rules
4.3.4 Correlation Analysis of Performance Indicators Considering Working Conditions and Dispatching Rules
4.3.5 Correlation Analysis Between Long-Term Performances and Short-Term Performances
4.3.6 Correlation Analysis of Performances on Production Line Named MIMAC
4.4 Correlation Analysis of Performances Based on Pearson Coefficient
4.4.1 Correlation Analysis of Daily WIP and Daily Moving Steps
4.4.2 Correlation Analysis of Daily Queue Leader and Daily Moving Steps
4.4.3 Correlation Analysis of Daily Equipment Utilization and Daily Moving Steps
4.5 Data Set of Performances for Semiconductor Manufacturing System
4.5.1 Training Set of Processing Cycle and Corresponding Short-Term Performances
4.5.2 Training Set of On-Time Delivery Rate and Corresponding Short-Term Performances
4.5.3 Training Set of Waiting Time and Corresponding Short-Term Performances
4.6 Summary
References
5 Data-Driven Release Control of Semiconductor Manufacturing System
5.1 Common Release Strategies of Semiconductor Manufacturing Systems
5.1.1 Common Release Control
5.1.2 Improved Release Control Strategy
5.1.3 Research Status of Release Control Strategies
5.2 Release Control Strategy Based on Extreme Learning Machine
5.2.1 Release Control Strategy for Determining Releasing Time Based on ELM
5.2.2 Release Control Strategy Based on Limit Learning Machine to Determine Releasing Sequence
5.3 Optimization of Release Control Based on Attribute Selection
5.3.1 Attribute Set Related to Releasing
5.3.2 Attribute Selection
5.3.3 Simulation Based on Attribute Selection
5.4 Summary
References
6 Data-Driving Dynamic Scheduling of Semiconductor Manufacturing System
6.1 Dynamic Dispatching Rules
6.1.1 Definition of Parameters and Variables
6.1.2 Hypothesis
6.1.3 Decision-Making Process
6.1.4 Simulation and Verification
6.2 Optimization of Algorithm Parameters Based on Data Mining
6.2.1 Overall Design
6.2.2 Algorithm Design
6.2.3 Process of Optimization
6.2.4 Simulation and Verification
6.3 Summary
References
7 Performance-Driving Dynamic Scheduling of Semiconductor Manufacturing System
7.1 Performance Prediction Method
7.1.1 Prediction Method of Long-Term Performances for Single-Bottleneck Semiconductor Production Line
7.1.2 Prediction Method of Long-Term Performances for Multi-bottleneck Semiconductor Production Line
7.2 Dynamic Scheduling of Semiconductor Production Line Based on Load Balancing
7.2.1 Overall Design
7.2.2 Load Balancing Technology
7.2.3 Selection of Parameters
7.2.4 Forecasting Model of Load Balancing
7.2.5 Dynamic Scheduling Algorithm Based on Load Balancing
7.2.6 Simulation and Verification
7.3 Dynamic Scheduling of Semiconductor Production Line Driven by Performances
7.3.1 Structure of Performance-Driving Scheduling Method
7.3.2 Dynamic Dispatching Rules
7.3.3 Prediction Model
7.3.4 Simulation and Verification
7.4 Summary
References
8 Development Trend of Scheduling Problems for Semiconductor Manufacturing System Under Big-Data
8.1 Industry 4.0
8.2 Industrial Big Data
8.3 Development Trend of Scheduling Problems for Semiconductor Manufacturing Under Big-Data
8.3.1 Data-Based Petri Net
8.3.2 Dynamic Simulation
8.3.3 Prediction Model
8.4 Application Example: Big Data Driving Forecasting Model in Complex Manufacturing System
8.5 Summary
References