✈️ 2015 US Flight Performance Analysis – Power BI Project

---

1. Project Objective

This project analyzes 5.8 million U.S. domestic flights from 2015 to evaluate operational performance across airlines, airports, and routes.

The objective is to identify delay patterns, reliability differences, and geographic concentration of disruptions using a structured, multi-page Power BI analytical report.

---

2. Dataset Overview

Source: US Flights Dataset (2015)

Total records analyzed: 5,819,079 flights

The dataset includes:

• Flight dates
• Airlines and IATA codes
• Origin and destination airports
• Departure & arrival delays (minutes)
• Cancellation and diversion indicators
• Route-level performance

A custom Date dimension table was created using DAX to enable full time-based analysis (Year, Month, Weekday hierarchy).

Data Model

The data model follows a structured star schema design:

• Fact Table: flights
• Dimension Tables: Date, airlines, airports_origin, airports_destination

This structure ensures scalable filtering, clean relationships, and optimized performance.

---

3. Dashboard Structure

📊 Page 1 – Executive Overview

Provides a high-level operational summary including:

• Total Flights
• Average Departure & Arrival Delay
• Delay Rate (15+ minutes)
• Cancellation Rate
• Diversion Rate
• Monthly delay trend

Purpose: Establish system-wide performance baseline for 2015.

---

✈️ Page 2 – Airline Performance Analysis

Compares airline reliability across:

• Average departure delay
• Delay rate (15+ minutes)
• Cancellation rate
• Volume vs delay relationship
• Weekday delay behavior among worst-performing carriers

Purpose: Identify performance dispersion and trade-offs between operational scale and delay severity.

---

🛫 Page 3 – Airport & Route Bottlenecks

Analyzes structural concentration of delays through:

• Geographic delay intensity map
• Top origin airports by average departure delay
• High-volume routes with elevated delay levels
• Route concentration share of total traffic

Purpose: Detect geographic and route-level pressure points in the flight network.

---

📈 Page 4 – Delay Patterns & Temporal Trends

Examines time-based behavior including:

• Average delay by weekday
• Month × weekday heatmap
• Monthly delay & cancellation trends
• Deviation from yearly baseline

Purpose: Identify recurring seasonal and weekday delay patterns.

---

4. Key Insights

• Average departure delay in 2015 was 9.3 minutes.
• 18.1% of flights experienced delays longer than 15 minutes.
• Delay levels increase during early summer and December.
• Mondays show the highest average weekday delays; Saturdays the lowest.
• Significant performance dispersion exists across airlines.
• Certain origin airports consistently exhibit elevated delay levels.
• High-volume routes account for a measurable share of total system traffic.
• Departure and arrival delays follow similar seasonal patterns but differ in magnitude.

Overall, the data suggests recurring operational pressure rather than purely random disruption events.

---

5. Business Relevance

Understanding delay concentration patterns allows airlines and airport operators to:

• Optimize scheduling buffers
• Identify congestion-prone hubs
• Improve route allocation strategies
• Prioritize operational improvements where impact is highest

This analysis transforms raw operational data into structured performance intelligence.

---

6. Technical Implementation

This project demonstrates:

• Star schema data modeling
• Custom Date dimension built using DAX
• Advanced DAX measures including:
  • Delay Rate (15+ min)
  • Cancellation Rate
  • Diversion Rate
  • Delay Deviation from Baseline
  • Top-N Route Concentration
• Context manipulation using ALL and REMOVEFILTERS
• Conditional formatting heatmap via Matrix visual
• Multi-page analytical storytelling
• Consistent KPI architecture across report pages

Tools Used:

• Power BI Desktop
• DAX
• Power Query

---