•Capacity Planning Reengineering: Leveraged multi-source data and AI/ML orchestration models to redesign capacity planning workflows, improving planning logic, responsiveness, and scalability across scenarios.

•Strategic Validation: Designed and executed A/B tests to compare alternative planning strategies, measuring uplift via key KPIs (e.g., forecast accuracy, utilization variance, service-level risk).

•Decision Support Systems: Built Power BI dashboards and internal websites to translate model outputs into executive-ready insights, clearer trade-offs, and more confident strategic decisions for executive leadership.

•Multi-Model Framework: Developed a robust forecasting framework integrating SARIMAX, Prophet, and GBR to enable long-term global capacity and demand planning.

•Advanced Segmentation: Applied Gaussian Mixture Models (GMM) for regional segmentation combined with advanced feature engineering.

•Impact: Successfully deployed models through an end-to-end CI/CD ML pipeline with auto-retraining capabilities, boosting forecast accuracy by 20%.

•Distributed ETL Systems: Designed and built scalable ETL pipelines using Python, Azure Data Factory, and Databricks.

•Big Data Processing: Leveraged Spark for TB-scale data processing and Airflow for automated workflow orchestration.

•Digital Transformation: drove the standardization and digital transformation of Intel’s global supply chain data infrastructure.

•Large-Scale Analysis: Utilized Python and SQL to analyze 10M+ rows of global procurement data to drive strategic sourcing decisions.

•Supplier Segmentation: Applied K-Means clustering for supplier segmentation and conducted A/B testing to validate strategies.

•Impact: Achieved a 10% cost reduction and a 15% reduction in procurement cycle time through data-driven optimization.

•Model Fine-Tuning: Fine-tuned Random Forest models on large-scale datasets to improve supplier selection accuracy.

•Decision Support: Enhanced sourcing decision quality by integrating predictive analytics into the supplier evaluation process.

•Power BI Optimization: Restructured Power BI schemas and optimized complex DAX queries to improve dashboard performance.

•Pipeline Engineering: Developed streamlined ETL pipelines for automated data flow and monitoring, significantly accelerating refresh speeds and operational efficiency.

•High-Precision Targeting: Analyzed 60M+ rows of user data, applying ML models (LR, GBDT, KNN, K-Means) and A/B testing to optimize multi-channel acquisition strategies.

•Engagement Growth: Successfully boosted ad targeting precision and user engagement by 15%.

•Campaign Analytics: Evaluated 9 major ad campaigns using multivariate and logistic regression, increasing operational efficiency by 34% while reducing costs by 15%.

•Inventory Optimization: Deployed advanced cloud-based SARIMAX models serving Dior’s operations across China, powering inventory management for 395 stores.

•Accuracy Improvement: Achieved a 10.5% boost in forecasting accuracy during high-volume promotional periods.

•Algorithm Tuning: Enhanced model performance via cross-validation (AUC, KS metrics), grid search, and feature engineering.

•Conversion Uplift: Resulted in a 28% increase in CTR (Click-Through Rate) and 8% in CVR (Conversion Rate) through iterative model refinement.

•Efficiency Boost: Created 30+ Python automation scripts to streamline manual supply chain processes.

•Time Savings: Saved 60+ work hours monthly, freeing up resources for strategic tasks.

•Digital Transformation: Advanced the department’s digital transformation through efficient data management practices and automated workflows.

•Anomaly Detection: Designed material consumption models and applied data mining on logistics and warehousing data to detect anomalies.

•Inventory Optimization: Addressed inventory imbalances through predictive analysis, streamlining supply chain operations.

•Impact: Raised customer satisfaction by 5% by ensuring smoother logistics and inventory availability.