📌 Project Overview

This project focuses on transforming a raw dataset into a structured model ready for analysis using Power BI.
While many dashboards are built from clean data, real-world datasets are often messy. The goal of this project was to demonstrate how I use Power Query and DAX in Power BI to clean, model, and prepare data for reporting.

---

🗂 Dataset

The dataset provided included inconsistencies such as:

• Mixed data types (numbers stored as text, dates unformatted)
• Missing/duplicate entries
• Unstructured flight/destination information

---

🔎 Transformation Process

1. Raw Dataset (Before)

Unstructured data with incorrect formats and duplicates.

---

2. Transformation Steps

A few key transformations performed in Power Query:

• Converted dep_time and arr_time (eg. 02:24) to time label format (eg. 'morning')
• Split airline details into airline / airplane
• Cleaned missing values and removed duplicates
• Standardized Date format
• Created calculated columns and measures using DAX

Example step:

I transformed the dep_time column into time-of-day categories by converting it to whole numbers, dividing with DAX, and applying IF-ELSE logic to label each period.

---

3. Cleaned Dataset (After)

Structured dataset ready for analysis and modeling.

---

📐 Data Model

Star schema built in Power BI, with fact and dimension tables connected.

---

📊 (Bonus) Summary Dashboard

A simple validation dashboard showing record counts and summary sales figures after cleaning.

---

🔎 Key Transformations & Insights

• Transformed dep_time into categorical labels using DAX & IF-ELSE logic, allowing segmentation by time of day.
• Merged the routes and flights tables to enrich flight records with route metadata.
• Grouped by destination, airline, and time of day, aggregating average duration and average price for each grouping.

📈 What This Enables

• Enables comparison of flight performance across different times of day and airlines.
• Helps identify which airline models and routes are more cost-effective based on average price vs. duration metrics.
• Lays groundwork for predictive analytics (e.g., predicting delays or optimizing routes) using clean, structured data.

---

🔑 Skills Demonstrated

• Data Cleaning & Transformation in Power BI (Power Query Editor)
• Handling inconsistent data types, duplicates, and missing values
• Merging multiple tables into a structured dataset
• Grouping and aggregating columns (e.g., average duration, average price)
• Building simple data models with relationships between tables
• Creating calculated columns and basic DAX measures
• Preparing datasets for future dashboarding and analysis
• Clear documentation of the data preparation process

---

🚀 How to Use

• Download the .pbix file from this repo.
• Open it with Power BI Desktop (free from Microsoft).

---