This project focuses on analyzing the flight activity and reservation processes of a major airline company. The aim of the project is to support decision-making by providing insights into the company's business processes.
The following are the business processes that this project focuses on:
Business Process | Granularity | Facts | Dimensions |
---|---|---|---|
Booking | Per selling ticket | Booking flight | Date, Passenger, Flight, Agency, Website, Promotion |
Frequent Flyer Program | Per transaction (upgrading, earning or redeeming miles, respond to promotions) | Loyalty fact | Date, Flight, FQ, Points redeeming key, Passenger |
Checking | Per inquiry | Factless table | Date, Passenger, Inquiry, Flight, Staff |
Upgrading | Per transaction (upgrading, earning or redeeming miles, respond to promotions) | Loyalty fact | Date, Flight, FQ, Points redeeming key, Passenger |
Cancelling | Per selling ticket | Booking flight | Date, Passenger, Flight |
Flight | Per flight | Flight | Date, Passenger, Airplane, Pilot, Crew, Flight, Route |
Landing/Takeoff | Per flight | Flight | Date, Passenger, Airplane, Flight, Route |
Transit | Per flight | Flight | Date, Passenger, Airplane, Flight, Route |
Booking Hotel | Per stay | Staying fact | Date, Flight, FQ, Hotel, Passenger |
Car Rent | Per transaction | Factless table | Date, Passenger, Flight |
Report Accident | Per incident | Factless table | Date, Flight |
Maintenance | Per airplane | Factless table | Airplane |
Report Inquiry | Per inquiry | Factless table | Date, Passenger, Inquiry, Flight, Staff |
Promote | Per transaction (upgrading, earning or redeeming miles, respond to promotions) | Loyalty fact | Date, Flight, FQ, Points redeeming key, Passenger |
Redeem Points | Per transaction (upgrading, earning or redeeming miles, respond to promotions) | Loyalty fact | Date, Flight, FQ, Points redeeming key, Passenger |
The project uses the Data Vault modeling approach due to the following reasons:
- Quick data extraction: The Data Vault model is faster, and the aviation industry operates around the clock.
- Structured format: The data is presented in a structured format.
- Easy integration: Sources can be easily integrated.
- Reduce 3NF complexity: The existing data links are typically modeled by a third normal form model, which might result in a solution that is rather rigid and requires a lot of rework as new sources are added.
- Historical management and raw data persistence: The Data Vault model is perfect for historical management and raw data persistence.
Dimension | Description |
---|---|
Date | Date dimension |
Passenger | Passenger dimension |
Flight | Flight dimension |
Airplane | Airplane dimension |
Pilot | Pilot dimension |
Crew | Crew dimension |
Agency | Agency dimension |
Website | Website dimension |
Promotion | Promotion dimension |
Hotel | Hotel dimension |
FQ | FQ dimension |
Inquiry | Inquiry dimension |
Staff | Staff dimension |
Points redeeming key | Points redeeming key dimension |
Fact | Description |
---|---|
Booking flight | Fact table for bookings |
Loyalty fact | Fact table for loyalty program transactions |
Staying fact | Fact table for hotel bookings |
Factless table | Factless table used for checking, car rent, report inquiry, report accident, and maintenance |
This project provides a comprehensive solution for analyzing the flight activity and reservation processes of a major airline company. The use of the Data Vault modeling approach provides a structured format for data presentation, easy integration, and historical management and raw data persistence.