DHL – Logistics Analytics Data Warehouse
Overview
This project consists of the development of a dimensional Data Warehouse model for logistics and sales analytics using a Star Schema architecture.
The solution was designed to support Business Intelligence operations related to:
- Sales analysis
- Shipment analytics
- Customer analysis
- Product return monitoring
- Profitability analysis
- Regional performance evaluation
The project integrates fact tables, dimension tables and analytical relationships optimized for Power BI reporting and KPI analysis.
Architecture
Star Schema Model
The data warehouse follows a classic Star Schema architecture composed of:
Fact Tables
| Table |
Description |
| FactSales |
Stores sales transaction metrics |
| FactReturns |
Stores product return information |
Dimension Tables
| Table |
Description |
| DimCustomer |
Customer information |
| DimProduct |
Product hierarchy and categories |
| DimDate |
Temporal analysis dimension |
| DimRegion |
Geographic information |
| DimShipMode |
Shipment methods |
Entity Relationship Structure
The schema connects transactional data with descriptive dimensions through foreign-key relationships optimized for analytical queries.
The model supports:
- Temporal analysis
- Geographic analysis
- Product analytics
- Customer segmentation
- Shipment analysis
- Return tracking
Fact Tables
FactSales
Main transactional sales table.
Attributes
| Attribute |
Type |
| RowID |
INT |
| Sales |
DECIMAL(10,2) |
| Quantity |
INT |
| Discount |
DECIMAL(3,2) |
| Order Priority |
VARCHAR(20) |
| OrderID |
VARCHAR(50) |
| Profit |
DECIMAL(10,2) |
| Shipping Cost |
DECIMAL(10,2) |
Foreign Keys
| Foreign Key |
References |
| ShipOrder |
DimShipMode |
| CodigoPostal |
DimRegion |
| DateKey_Ship |
DimDate |
| DateKey_Order |
DimDate |
| CustomerID |
DimCustomer |
| ProductID |
DimProduct |
FactReturns
Stores returned product transactions.
Attributes
| Attribute |
Type |
| ReturnID |
INT |
| ReturnQuantity |
INT |
| ReturnAmount |
DECIMAL(10,2) |
| OrderID |
VARCHAR(50) |
| Reason |
VARCHAR(50) |
Foreign Keys
| Foreign Key |
References |
| CodigoPostal |
DimRegion |
| DateKey_Order |
DimDate |
| CustomerID |
DimCustomer |
Dimension Tables
DimCustomer
| Attribute |
| CustomerID |
| CustomerName |
| Segment |
| CodigoPostal |
| IsCurrent |
DimProduct
| Attribute |
| ProductID |
| ProductName |
| Category |
| SubCategory |
DimDate
| Attribute |
| DateKey_Order |
| DateKey_Ship |
| Year |
| Date |
| Semester |
| Week |
| Quarter |
| Month |
| Weekday |
DimRegion
| Attribute |
| CodigoPostal |
| RegionName |
| CountryName |
| StateName |
| Market |
| CityName |
DimShipMode
| Attribute |
| ShipOrderID |
| ShipModeName |
| Method |
Business Objectives
The model was designed to answer analytical questions such as:
- Which products generate the highest sales?
- Which regions generate the highest profit?
- Which customers spend the most?
- What is the impact of discounts on profit?
- Which shipment methods are most used?
- What are the main return reasons?
- How do sales evolve over time?
Technologies Used
| Technology |
Purpose |
| Power BI |
Dashboard visualization |
| Excel |
Source data |
| SQL Concepts |
Data modeling |
| Star Schema |
Data warehouse architecture |
| DAX |
KPI calculations |
| Power Query |
Data transformation |
Data Warehouse Workflow
Analytical Capabilities
Sales Analytics
- Revenue analysis
- Profitability tracking
- Product performance
- Customer segmentation
Logistics Analytics
- Shipment mode analysis
- Delivery analysis
- Shipping cost monitoring
Return Analytics
- Product return tracking
- Return reason analysis
- Return quantity monitoring
Geographic Analytics
- Regional profit analysis
- Country-level analysis
- Market segmentation
Advantages of the Model
| Advantage |
Description |
| Performance |
Optimized analytical queries |
| Scalability |
Easy integration of new dimensions |
| Simplicity |
Easy BI dashboard integration |
| Flexibility |
Supports multiple business analyses |
| Maintainability |
Clear separation of dimensions and facts |
Requirements
| Requirement |
Description |
| Power BI Desktop |
Dashboard development |
| Excel |
Data source handling |
| SQL Knowledge |
Optional for extensions |
| Windows OS |
Recommended environment |
Future Improvements
- SQL Server migration
- ETL pipeline automation
- Real-time analytics
- Incremental data refresh
- Cloud data warehouse integration
- Predictive analytics
- Machine learning integration
Disclaimer
This project was developed for academic and Business Intelligence purposes.
Users remain responsible for validating data integrity and analytical outputs before operational usage.
Author
José Garcia
Data Scientist
Business Intelligence & Data Warehouse Engineering