This repository hosts a comprehensive data science project conducted at Kalbe Nutritionals. The project addresses critical challenges faced by both the inventory and marketing teams. It involves predictive modeling for daily sales quantity and customer segmentation. The goal is to provide insights and solutions to optimize inventory management and enhance marketing strategies.
- Data Ingestion: Utilized PostgreSQL and DBeaver for data ingestion, and performed exploratory data analysis using SQL queries.
- Interactive Dashboards: Ingested data into Tableau Public to create interactive dashboards, providing insights into daily sales performance, monthly quantity trends, product sales by quantity, and sales performance by store.
- Predictive Modeling: Employed Python in Google Colab, focusing on time series ARIMA for predictive modeling. Ensured data cleansing and preprocessing for accurate daily sales quantity predictions for all Kalbe products.
- Clustering: Utilized clustering techniques, including the elbow method, silhouette plots, and K-means clustering, to segment customers for personalized marketing strategies.
To get started with this project, follow these steps:
- Clone the repository to your local machine.
- Review the project notebooks to understand the data analysis, predictive modeling, and clustering processes.
- Use the provided SQL queries for data analysis with DBeaver and PostgreSQL.
- Explore the Tableau dashboards to gain insights into sales performance and customer segmentation.
- Python
- Google Colab
- Tableau Public
- DBeaver
- PostgreSQL