Maven Airlines Challenge

After taking part in my first data analytics challenge hosted by Maven Analytics (The Maven Unicorns Challenge), I was hungry to work on more projects. I wanted to challenge myself and work on my Power BI skills and that’s when I found out about this challenge on the Maven Analytics LinkedIn page.

Objective

To understand the passenger satisfaction ratings of services rendered by Maven Airlines (A hypothetical airline company) KPIs to be delivered:

• Which percentage of airline passengers are satisfied? Does it vary by customer type? What about the type of travel?
• What is the customer profile for a repeating airline passenger?
• Which factors contribute to customer satisfaction the most? What about dissatisfaction?

Here is the breakdown of my thought process

Data Extraction Process:

I have used SQL Server Management Studio (SSMS) to import and clean the data for this project.

Data Cleaning and Validation Process:

After importing the dataset known as ‘Services_Data Table’ and storing it in a database, changes related to replacing values and adding new columns were executed to improve the data analysis process. In addition, another table called 'Airline Services' comprises two columns, namely Serial No. and Services offered by the Airlines were created.

Data Modelling Process:

Having made all the changes, both tables were loaded into Power BI using the Direct Query feature. In the model view of Power BI, the two tables, namely the Airline Services Table and the Services_Data Table are connected by creating a one-to-many relationship with a one-way filter from the ID column of the Airline Services Table to the ID column of Services_Data Table.

Data Analysis & Visualization Process:

A few measures and data groups were created to analyze the KPIs. To improve the analysis, I visualized these measures and data groups in the form of cards, charts and graphs set against some filters to regulate the flow of data. The insights and recommendations for the KPIs have been elaborated in the dashboard.

The static images of the dashboard are revealed below. (Click on the image to view it.)

Your thoughts and feedback on this are much appreciated.