Airline Flight Performance Analysis

Key Metrics

98,619

Total Records

33.3%

On-Time Rate

33.3%

Delay Rate

33.4%

Cancellation Rate

Project Overview

About the Dataset

Dataset Information

Dataset: Synthetic Airline Dataset (Kaggle)
Author: Sourav Banerjee
Tool: Tableau Public
Records: 98,619 passenger-flight records
Target: Flight Status
Categories: On Time, Delayed, Cancelled

Business Context

Airlines generate vast amounts of passenger and flight data, yet delays and cancellations remain critical operational and customer experience challenges. Understanding the factors behind flight disruptions is essential for improving efficiency, resource allocation, and passenger satisfaction.

Business Question: "What demographic, geographic, and temporal factors are associated with flight delays and cancellations?"

Methodology

STAR Framework

S

Situation

Airlines need better visibility into disruptions such as delays and cancellations to improve operational planning and customer experience.

T

Task

Analyze airline data and build an interactive dashboard to identify flight status patterns and key contributing factors.

A

Action

Cleaned and profiled the data, created calculated fields, standardized continent names, and designed a comprehensive Tableau dashboard.

R

Result

Delivered an interactive dashboard summarizing flight performance and highlighting patterns by time, geography, airport, age group, gender, and nationality.

Live Dashboard

Interactive Tableau Dashboard

Explore flight performance data interactively. Analyze delays, cancellations, and on-time performance by airport, continent, passenger demographics, and time.

Open dashboard in full screen

Project Presentation

Presentation Slides

A detailed walkthrough of the analysis, methodology, and key findings.

Open Presentation

Findings

Key Insights

98,619 records analyzedThe dataset contains nearly 100K passenger-flight records covering multiple airlines, airports, and time periods.

Balanced target distributionFlight status categories are almost evenly split: On Time (33.3%), Delayed (33.3%), Cancelled (33.4%).

Europe - highest delay rateEurope showed the highest delay rate at approximately 33.9% among all continents.

South America - highest cancellation rateSouth America's cancellation rate reached approximately 34.0%, the highest across continents.

Gender-based parityPerformance metrics by gender were very similar, with no meaningful imbalance in delay or cancellation rates.

Nationality patterns (with caution)Some nationalities show high delay rates, but many had very small record counts - interpret with care.

Limitations

The dataset is synthetic and does not represent real airline operations.
The target variable (Flight Status) is artificially balanced - not typical of real-world distributions.
Results should be interpreted as analytical patterns, not real operational performance.
Some categories (e.g., specific nationalities) have very small record counts, which may produce misleading high rates.

Future Improvements

Add minimum-volume filters for nationalities and airports to avoid sparse-data artifacts.
Include route-level analysis to identify high-risk origin-destination pairs.
Incorporate weekday and holiday effects on flight status.
Build predictive models for flight status classification.
Evaluate models using accuracy, F1-score, and confusion matrix.