VIP20
本课程属于：006-Data Science and Engineering Analytics English Technical Courses · 共 20 门相关课程，组合学习更完整。
查看所属VIP全部课程 →
← 返回所属目录 咨询 / 购买

Data Science for Manufacturing Yield Optimization

30 Chapters
View Course Samples Online
01
Yield optimization in manufacturing
definitions, key metrics, and the data science opportunity.
optimizationmanufacturing
02
Semiconductor and discrete manufacturing process flows
critical steps, yield loss mechanisms, and data generation points.
Semiconductordiscrete
03
Data acquisition and sensor integration
SCADA, MES, FDC, and metrology data sources for yield analysis.
acquisitionintegration
04
Data cleaning and preprocessing for manufacturing data
missing values, outliers, time alignment, and lot tracking.
cleaningpreprocessing
05
Exploratory data analysis for yield
wafer maps, spatial patterns, and defect density visualization.
Exploratoryanalysis
06
Statistical process control and yield monitoring
control charts, Cpk, and Western Electric rules.
Statisticalmonitoring
07
Feature engineering for yield prediction
derived process metrics, windowed statistics, and interaction terms.
predictionwindowed
08
Correlation analysis and root cause screening
identifying process parameters linked to yield loss.
Correlationanalysis
09
Regression models for yield prediction
linear, regularized, and generalized linear models.
Regressionprediction
10
Tree-based ensemble methods for yield classification
random forests, gradient boosting, and XGBoost.
Tree-basedensemble
11
Handling class imbalance in yield failure prediction
resampling, cost-sensitive learning, and anomaly detection.
Handlingimbalance
12
Time-series analysis of process drift
trend detection, change point analysis, and autocorrelation.
Time-seriesanalysis
13
Spatial yield modeling
wafer-level and die-level spatial statistics, kriging, and spatial regression.
modelingwafer-level
14
Multivariate analysis for fault detection and diagnosis
PCA, PLS, and contribution plots.
Multivariateanalysis
15
Clustering and segmentation of process runs
identifying distinct yield loss signatures.
Clusteringsegmentation
16
Deep learning for yield
convolutional neural networks for wafer map defect pattern recognition.
convolutionalnetworks
17
Sequence models for process trace data
RNNs, LSTMs, and transformers for time-series yield prediction.
SequenceRNNs
18
Model evaluation and validation for yield models
cross-validation, temporal splits, and business metrics.
evaluationvalidation
19
Interpretability and explainability in yield models
SHAP, LIME, and feature importance for engineering action.
Interpretabilityexplainability
20
Design of experiments and response surface methodology for yield optimization.
Design of experiments and response surface methodology for yield optimization.
experimentsresponse
21
Bayesian optimization and active learning for process parameter tuning.
Bayesian optimization and active learning for process parameter tuning.
Bayesianoptimization
22
Reinforcement learning for adaptive process control and yield improvement.
Reinforcement learning for adaptive process control and yield improvement.
Reinforcementadaptive
23
Virtual metrology and soft sensing
predicting quality from process data to reduce metrology steps.
metrologypredicting
24
Predictive maintenance and its impact on yield
equipment health, chamber matching, and drift compensation.
Predictivemaintenance
25
Yield ramp and new product introduction
transfer learning and few-shot learning for limited data.
transferfew-shot
26
Integration of yield analytics into MES and APC systems
deployment, APIs, and real-time inference.
Integrationanalytics
27
Data pipeline and MLOps for manufacturing
versioning, monitoring, and model retraining.
pipelineMLOps
28
Case study
end-to-end yield optimization in semiconductor fabrication.
end-to-endoptimization
29
Case study
yield improvement in electronics assembly and PCB manufacturing.
improvementelectronics
30
Future directions
AI-driven autonomous yield management, digital twins, and edge analytics.
directionsAI-driven