Project Highlights
Demand Forecasting for Independent Contractors
- Developed ML models to predict workforce demand across 168 warehouses
- Utilized Random Forest algorithm for superior performance
- Enabled data-driven decision making for workforce management
Order Count Prediction for Columbia Sportswear
- Implemented time series forecasting models (Prophet, ARIMA)
- Conducted seasonal analysis using Random Forest and Linear Regression
- Improved operational cost estimation and service planning
Technologies & Skills
- Cloud Platforms: Azure, Databricks
- Programming: Python, SQL, C#
- Machine Learning: Time Series Forecasting, Random Forest, Linear Regression
- Big Data: Apache Spark, ETL pipelines
- DevOps: Microsoft SC360, Azure Pipelines
- Data Engineering: Delta Live Tables, Data Warehousing
Key Achievements
- Successfully migrated data from Softeon - USPZN3 to Symphony application
- Developed SQL Backend Procedure and API for a C# & .NET framework web app
- Implemented end-to-end ML pipelines from data extraction to model deployment
- Contributed to agile development processes and sprint planning
Impact
The machine learning models developed during this internship enabled UPS to:
- Optimize workforce allocation across warehouses
- Improve cost estimation and service planning for key customers
- Enhance data-driven decision making in supply chain operations