Airline Data Analysis Dashboard

A comprehensive data analysis project for CS699 (Software Lab) under Prof. Bhaskaran Raman, focusing on airline passenger data visualization and insights.

📊 Project Overview

This project analyzes airline passenger data to provide insights into travel patterns, demographics, and operational efficiency. The system includes a Python-based backend for data processing and visualization generation, coupled with a web-based frontend dashboard for interactive data exploration.

🎯 Features

• Comprehensive Data Analysis: Analysis of 98,619 passenger records with automated pipeline
• Rich Visualizations: 21 graphs and charts with meaningful descriptive names
• Demographic Insights: Gender, age, nationality, and geographic distribution analysis
• Flight Performance Metrics: On-time, delayed, and cancelled flight statistics
• Web Dashboard: User-friendly interface with 23 pages (21 graphs + index + additional pages)
• Multi-continent Analysis: Dedicated bar & pie charts for 6 continents (Asia, Europe, North America, South America, Africa, Oceania)
• Progress Tracking: Real-time feedback during script execution with 5-phase progress system
• YAML Configuration: Centralized config.yaml for easy customization of paths and parameters
• Export Capabilities: High-resolution PNG images (300 DPI) compiled in PDF format

🛠️ Tech Stack

Backend

• Python 3.10+
• NumPy (≥1.21.0) - Numerical computing
• Pandas (≥1.3.0) - Data manipulation and analysis
• Matplotlib (≥3.4.0) - Static plotting
• Seaborn (≥0.11.0) - Statistical data visualization
• Plotly (≥5.0.0) - Interactive visualizations
• Kaleido (≥0.2.1) - Static image export for Plotly
• PyYAML (≥6.0) - Configuration file parsing
• TQDM (≥4.64.0) - Progress bars for script execution

Frontend

• HTML5 - Structure and content
• CSS3 - Styling and responsive design
• JavaScript - Interactive elements

📁 Project Structure

Airline_Record_Analysis/
├── README.md                # Project documentation
├── requirements.txt         # Python dependencies
├── .gitignore              # Git ignore rules
├── config.yaml             # Configuration file for paths and settings
├── venv/                   # Python virtual environment
├── backend/
│   ├── main.py             # Main data analysis script (319 lines)
│   ├── data/
│   │   └── Airline_Dataset.csv  # Raw dataset (98,621 records)
│   └── graphs/             # Generated visualization images (21 graphs)
│       ├── gender_distribution_pie.png           # Gender distribution
│       ├── top_countries_passengers_bar.png      # Top countries by passengers
│       ├── continent_country_sunburst.png        # Continent hierarchy
│       ├── age_distribution_histogram.png        # Age histogram
│       ├── age_count_distribution_bar.png        # Age count distribution
│       ├── nationality_distribution_pie.png      # Nationality analysis
│       ├── flight_status_distribution_pie.png    # Flight status overview
│       ├── flight_status_trends_line.png         # Flight trends over time
│       ├── average_age_trends_line.png           # Age trends over time
│       ├── asia_passengers_bar.png               # Asia passenger bar chart
│       ├── asia_passengers_pie.png               # Asia passenger pie chart
│       ├── europe_passengers_bar.png             # Europe passenger bar chart
│       ├── europe_passengers_pie.png             # Europe passenger pie chart
│       ├── north_america_passengers_bar.png      # North America bar chart
│       ├── north_america_passengers_pie.png      # North America pie chart
│       ├── oceania_passengers_bar.png            # Oceania passenger bar chart
│       ├── oceania_passengers_pie.png            # Oceania passenger pie chart
│       ├── africa_passengers_bar.png             # Africa passenger bar chart
│       ├── africa_passengers_pie.png             # Africa passenger pie chart
│       ├── south_america_passengers_bar.png      # South America bar chart
│       └── south_america_passengers_pie.png      # South America pie chart
└── frontend/
    ├── index.html          # Main dashboard page
    ├── style.css           # Dashboard styling
    ├── img/                # Generated graph visualizations (copied from backend/graphs)
    │   ├── gender_distribution_pie.png           # Gender distribution
    │   ├── top_countries_passengers_bar.png      # Top countries
    │   ├── continent_country_sunburst.png        # Continent hierarchy
    │   ├── age_distribution_histogram.png        # Age histogram
    │   ├── age_count_distribution_bar.png        # Age count
    │   ├── nationality_distribution_pie.png      # Nationality
    │   ├── flight_status_distribution_pie.png    # Flight status
    │   ├── flight_status_trends_line.png         # Flight trends
    │   ├── average_age_trends_line.png           # Age trends
    │   └── (Regional bar & pie charts for 6 continents)
    ├── static/             # Static assets for UI
    │   ├── dataset_img.png # Dataset preview icon
    │   └── pdf_img.png     # PDF report icon
    ├── pages/              # Individual graph pages (23 HTML files)
    │   ├── Graph1.html through Graph23.html
    │   └── Each page displays a specific visualization
    └── reports/            # Project deliverables
        ├── Airline_Dataset.csv      # Dataset copy
        ├── generated-graphs.pdf     # Compiled visualizations
        ├── project-proposal.pdf     # Project proposal document
        ├── project-update.pdf       # Project update document
        └── 23m0761_23m0773.tar.xz  # Compressed project archive

📈 Data Analysis Insights

The project generates 21 comprehensive visualizations with meaningful, descriptive filenames including:

1. Demographic Analysis

   • Gender distribution of passengers (pie chart)
   • Age distribution histogram and trends
   • Nationality breakdown

2. Geographic Analysis

   • Top 10 countries by passenger volume (bar chart)
   • Continent-wise passenger distribution (sunburst chart)
   • Regional travel patterns across 6 continents

3. Flight Operations

   • Flight status analysis (On-time, Delayed, Cancelled)
   • Temporal trends in flight performance
   • Departure date patterns

4. Regional Deep-dives

   • Asia: Passenger analysis with bar and pie charts
   • Europe: Travel patterns visualization
   • North America: Market insights
   • South America: Passenger distribution
   • Africa: Regional breakdown
   • Oceania: Travel statistics

5. Advanced Analytics

   • Quadrant analysis for multi-dimensional insights
   • Age trends with minimum threshold filtering
   • Hierarchical data representation

🚀 Getting Started

Prerequisites

• Python 3.10+
• All required packages with versions listed in requirements.txt

Configuration

The project uses a config.yaml file to manage all file paths and analysis settings. This centralized configuration makes it easy to modify input/output locations and analysis parameters without changing the code.

Important: All paths in config.yaml are relative to the project root directory. Always run the scripts from the project root:

Airline_Record_Analysis$ python3 backend/main.py

Key Configuration Sections:

• Input Configuration: Dataset path and date parsing settings
• Output Configuration: Directory for generated graphs and individual graph filenames
• Figure Settings: DPI and figure sizes for visualizations
• Analysis Settings: Parameters like top N countries, nationalities, and age bins

To modify paths or settings, edit the config.yaml file in the project root:

input:
  dataset_path: "backend/data/Airline_Dataset.csv"
  parse_dates: ["Departure Date"]

output:
  graphs_dir: "backend/graphs"
  graphs:
    gender_distribution: "gender_distribution_pie.png"
    top_countries: "top_countries_passengers_bar.png"
    continent_country_sunburst: "continent_country_sunburst.png"
    age_distribution: "age_distribution_histogram.png"
    age_count_distribution: "age_count_distribution_bar.png"
    # ... (21 total graph definitions with meaningful names)

figure_settings:
  dpi: 300
  large_figure_size: [14, 8]
  xlarge_figure_size: [18, 10]

analysis:
  top_n_countries: 10
  top_n_nationalities: 20
  age_bins: 20

Running the Analysis

1. Clone the repository:

   git clone https://github.com/rohitsingh25/Airline_Record_Analysis.git
   cd Airline_Record_Analysis

2. Create and activate virtual environment:

   # Create virtual environment
   python3 -m venv venv
   
   # Activate virtual environment
   # On Linux/Mac:
   source venv/bin/activate
   
   # On Windows:
   # venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Run the backend analysis (from project root):

   python3 backend/main.py

   This will:

   • Load configuration from config.yaml
   • Read the dataset from backend/data/Airline_Dataset.csv (98,619 records)
   • Display 5-phase progress with status updates:
     • Phase 1: Configuration & Data Loading
     • Phase 2: Data Preprocessing & Validation
     • Phase 3: Demographic Analysis (3 visualizations)
     • Phase 4: Flight Operations Analysis (3 visualizations)
     • Phase 5: Regional Analysis (12 visualizations)
   • Generate all 21 visualizations with meaningful names
   • Save graphs to backend/graphs/ directory as high-resolution PNG files (300 DPI)
   • Copy all graphs to frontend/img/ directory for dashboard display

5. View the dashboard:

   • Open frontend/index.html in your web browser directly, or
   • From project root, run:

     # On Linux/Mac:
     xdg-open frontend/index.html
     
     # On Mac:
     open frontend/index.html
     
     # On Windows:
     start frontend/index.html

   • Navigate through 23 different pages including index and 21 visualization graphs
   • Each graph page displays one of the 21 generated visualizations with meaningful names
   • Explore continent-specific analyses (6 continents with bar & pie charts each)
   • View demographic insights, flight operations, and regional patterns
   • Access project reports and dataset from the frontend/reports/ directory

6. Deactivate virtual environment (when done):

   deactivate

Dataset Information

The airline dataset (Airline_Dataset.csv) contains:

• 98,621 passenger records
• Passenger demographics: ID, Name, Gender, Age, Nationality
• Flight details: Departure/Arrival airports, dates, pilot information
• Geographic data: Country codes, continents, airport locations
• Operational data: Flight status (On Time, Delayed, Cancelled)

Available in two locations:

• backend/data/Airline_Dataset.csv - Source data for analysis
• frontend/reports/Airline_Dataset.csv - Reference copy for reporting

📊 Project Deliverables

The frontend/reports/ directory contains:

• generated-graphs.pdf - Compilation of all 21 visualizations
• project-proposal.pdf - Initial project proposal document
• project-update.pdf - Project progress update
• 23m0761_23m0773.tar.xz - Complete project archive

🔍 Key Findings

• Analysis covers 98,619 passengers from 6 continents with diverse demographics
• 21 high-resolution visualizations (300 DPI) covering demographic, geographic, and operational aspects
• Meaningful graph naming convention improves navigation and understanding
• Flight performance metrics reveal operational efficiency patterns across time periods
• Temporal analysis shows trends in passenger age and flight status
• Geographic distribution highlights major travel corridors and regional patterns
• Continent-specific insights: 2 visualizations (bar & pie) for each of 6 continents
• Clean web dashboard with organized pages, responsive design, and intuitive navigation
• Centralized YAML configuration enables easy customization of analysis parameters

👥 Contributors

• Rohit Singh Yadav - LinkedIn
• Bharat Patidar - LinkedIn

📧 Contact

For questions or collaboration opportunities, please reach out through our LinkedIn profiles or create an issue in this repository.

📄 License

This project is part of academic coursework for CS699 (Software Lab) and is intended for educational purposes..

---

CS699 Software Lab Project | IIT Bombay