An end-to-end BMW sales analytics and machine learning project using Excel, Power Query, Power Pivot, Python, and Scikit-learn.
The project combines:
- Business Intelligence
- Data Analysis
- Interactive Dashboarding
- Machine Learning Classification
This project analyzes BMW global sales data from 2010 to 2024 and predicts whether sales are classified as High Sales or Low Sales using KNN Classification.
The solution includes:
- Interactive Excel Business Intelligence Dashboard
- Data Cleaning & Transformation
- Machine Learning Pipeline
- Business KPI Analysis
- Microsoft Excel
- Power Query
- Power Pivot
- Pivot Tables
- Data Visualization
- KPI Analysis
- Python
- Pandas
- NumPy
- Scikit-learn
- KNN Classification
- Joblib
- Total Revenue Analysis
- Sales by Region
- Fuel Type Distribution
- Transmission Revenue Analysis
- Sales Volume by Model
- Interactive Filters & Slicers
- KPI Monitoring
- Removed inconsistencies
- Validated missing values
- Standardized categories
- Created calculated columns
- Revenue calculations
- Sales classification
- KPI extraction
- Regional aggregations
Sales were classified into:
- High Sales
- Low Sales
using threshold-based classification and analytical calculations.
- KNN Classification
Predict:
- High Sales
- Low Sales
- Data Cleaning
- Feature Encoding
- Feature Scaling
- Model Training
- Model Serialization (.pkl)
├── Dashboard/
├── Data/
├── Machine_Learning/
├── README.md
└── requirements.txt
- Asia generated the highest regional sales revenue.
- Automatic transmission vehicles contributed significantly to total revenue.
- Fuel type distribution remained balanced across categories.
- Certain BMW models consistently dominated sales volume.
- Deploy ML model using Flask/FastAPI
- Real-time dashboard integration
- Advanced predictive analytics
- API-based data ingestion
- Cloud deployment
Kareem Abdelrhman
GitHub: https://github.com/Kareem-opps