Business Objective: Predict DoorDash delivery duration accurately to enhance customer experience and optimize delivery operations.
Approach: We have employed a comprehensive approach to build a robust delivery duration prediction system for DoorDash. This approach includes:
- Utilizing a combination of regression models
- Performing extensive feature engineering
- Employing stacking ensemble techniques
Tools Used: We leveraged various tools and libraries in Python to accomplish this project, including:
- Python
- Pandas
- Scikit-Learn
- LightGBM
- XGBoost
- RandomForest
- Ridge Regression
- GridSearchCV
By combining these tools and techniques with historical data, we aim to provide DoorDash with an accurate and efficient system for predicting delivery durations, ultimately improving customer satisfaction and optimizing delivery operations.