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

Causal Inference for Industrial Process Improvement

30 Chapters

View Course Samples Online

Foundations of causal inference in industrial process improvement

correlation vs. causation, potential outcomes framework, and the role of randomization in manufacturing and production systems.

Foundationsinference

Causal graphs and structural causal models for process engineering

directed acyclic graphs (DAGs), confounding, colliders, mediators, and backdoor paths in industrial settings.

structuraldirected

Identification strategies for observational process data

backdoor adjustment, frontdoor adjustment, instrumental variables, and do-calculus with examples from quality control and process optimization.

Identificationstrategies

Randomized experiments in industrial processes

design of experiments (DOE), factorial and fractional factorial designs, blocking, and randomization constraints in production environments.

Randomizedexperiments

Propensity score methods for process improvement

matching, stratification, inverse probability weighting, and their application to observational studies in manufacturing and operations.

Propensityimprovement

Regression discontinuity designs in industrial contexts

sharp and fuzzy RDD, threshold-based process changes, and applications to quality inspection and maintenance policies.

Regressiondiscontinuity

Difference-in-differences and synthetic control methods for process interventions

panel data, parallel trends assumption, and evaluating plant-level or line-level improvements.

Difference-in-differencessynthetic

Instrumental variables and natural experiments in industrial systems

finding valid instruments, two-stage least squares, and applications to supply chain and production interventions.

Instrumentalvariables

Mediation analysis in process improvement

direct and indirect effects, causal mediation, and understanding mechanisms in quality and productivity interventions.

Mediationanalysis

Heterogeneous treatment effects and causal machine learning

causal forests, meta-learners, and personalized process adjustments for varying production conditions.

Heterogeneoustreatment

Sensitivity analysis and robustness checks for causal claims in industrial settings

unmeasured confounding, E-values, and stress-testing conclusions.

Sensitivityanalysis

Time series causal inference for industrial processes

Granger causality, transfer entropy, and causal impact analysis for continuous and batch processes.

inferenceindustrial

Bayesian causal inference for process improvement

Bayesian networks, probabilistic graphical models, and decision-making under uncertainty in manufacturing.

Bayesianinference

Causal discovery algorithms for process data

constraint-based, score-based, and functional causal models for learning causal structure from observational process data.

discoveryalgorithms

reducing defects in semiconductor manufacturing using causal inference to identify root causes and optimize process parameters.

reducingsemiconductor

improving yield in chemical production through causal analysis of reactor conditions and raw material variability.

improvingchemical

optimizing energy consumption in industrial facilities using causal inference to separate effects of operational changes from external factors.

optimizingconsumption

evaluating maintenance interventions in manufacturing using causal methods to estimate impact on downtime and equipment reliability.

evaluatingmaintenance

supply chain process improvement with causal inference to assess supplier changes and logistics interventions on delivery performance.

improvementinference

quality control in pharmaceutical manufacturing using causal inference for process validation and continued process verification.

pharmaceuticalmanufacturing

Practical implementation

data requirements, preprocessing, and software tools (Python, R, causal inference libraries) for industrial causal analysis.

implementationrequirements

Designing and analyzing A/B tests in industrial digital systems

online experiments, switchback tests, and challenges in production environments.

Designinganalyzing

Causal inference for process monitoring and fault detection

integrating causal models with statistical process control (SPC) and anomaly detection.

inferencemonitoring

Combining causal inference with optimization

using causal estimates to inform response surface methodology and process setpoint optimization.

Combininginference

Causal inference in the presence of interference and spillover effects

network effects in production lines and supply chains.

inferencepresence

Handling missing data and measurement error in causal analysis for industrial processes

multiple imputation, sensitivity bounds, and error-in-variables.

Handlingmeasurement

Ethical and organizational considerations for causal inference in industry

data privacy, algorithmic bias, and change management for evidence-based process improvement.

organizationalconsiderations

Advanced topics

continuous treatments, dose-response functions, and causal inference for complex interventions in industrial systems.

Advancedcontinuous

Integrating causal inference with machine learning and AI for industrial process improvement

hybrid models, explainability, and decision support systems.

Integratinginference

Capstone project

end-to-end causal inference study for a real or simulated industrial process improvement problem, from problem definition to actionable recommendations.

Capstoneend-to-end