An end-to-end analytical review of US domestic flight operations — diagnosing systemic inefficiencies across delays, cancellations, reliability, economic cost, and environmental impact.
This project transforms millions of flight records into decision-ready insights for airlines, airports, regulators, and operations teams. The goal is not just reporting but diagnosing systemic operational inefficiencies and highlighting where performance, reliability, cost control, and sustainability can be improved.
Modern aviation systems are highly interconnected — a delay at one airport or route can cascade across an entire network. This project explores how, when, where, and why these disruptions occur across network-wide operational performance, root causes of delays, airline and airport reliability benchmarking, route-level risk identification, and financial and environmental externalities of inefficiency.
The analysis spans 10 years and over 5.6 million flights, enabling both trend analysis and structural performance comparisons.
Critical findings across system performance, operational efficiency, reliability, infrastructure, financial cost, and sustainability.
Avg arrival delay (min) by day of week · 33.88% of all flights delayed
Reliability score (%) · 14-point spread between best and worst
Highest cancellation rates by airport
Regional airports show 15–25% cancellation rates vs. 2.78% national average
Avg delay (min) on worst routes · all at 100% delay rate
Minutes average delay · systematic failures requiring network redesign
2024 peak · cost by time-of-day window
2024 peak · top emitting airports (M metric tons)
Smaller airports show disproportionately high cancellation rates (15–25%), indicating service reliability challenges.
14-point spread between best (Hawaiian 86%) and worst (JetBlue 72%) performers indicates operational excellence opportunities.
Late evening operations accumulate delays throughout the day, suggesting schedule compression issues.
CO₂ emissions from delays represent significant sustainability risk and potential regulatory exposure.
Routes with 100% delay rates indicate systematic problems requiring network redesign.
Eight focused analytical dashboards organized for stakeholders at every level — from executive KPIs to granular route-level diagnostics. Each page is purpose-built to answer a specific set of business questions.
The entry point for the entire report. This page presents the system-wide health of US domestic aviation at a glance — total flights, on-time performance rate, cancellation rate, and average departure and arrival delays. KPI cards give leadership an immediate read on whether operations are improving or deteriorating, while summary visuals break down completed vs. cancelled flights and year-over-year performance trends.
This page diagnoses why delays happen and when they concentrate. Late aircraft (33.73%) and carrier delays (32.41%) together account for two-thirds of all disruptions. Time-of-day and day-of-week patterns expose Monday peaks and the COVID-era dip in 2020–2021, followed by a sharp post-recovery surge peaking at 22 min average in 2024.
A ranking-based benchmarking page evaluating every airport by average arrival delay and cancellation rate. RANKX-driven measures surface regional bottlenecks — Morgantown (25.81% cancellation) and Watertown (24.19%) carry disproportionate risk. USERELATIONSHIP toggles between origin and destination perspectives.
Benchmarks carriers using a composite reliability score balancing on-time performance against cancellation behavior. The 13-point gap between Hawaiian (85.45%) and JetBlue (72.45%) is immediately visible, giving regulators and passengers a data-backed view of which carriers deliver consistent service.
The most granular operational page. It evaluates origin–destination pairs by average delay, delay rate, and cancellation frequency. Routes like CAK→TYS (1,237 min, 100% delay rate) and MDT→HPN (798 min) surface as clear candidates for network redesign. Route identifiers are feature-engineered from origin and destination codes.
Translates delay minutes into estimated dollar cost using industry per-minute cost factors. The 7–10 PM window accounts for the highest concentration ($0.60–$0.62bn per hour slot). 2024 peaked at $1.21 billion — converting an abstract operational metric into a tangible business case for schedule optimization.
Estimates excess fuel burn and CO₂ emissions directly attributable to delay minutes. Chicago O’Hare leads at 10.1M tonnes, followed by Atlanta (8.6M), DFW (7.8M), and Denver (7.4M). 2024 peaked at 26M tonnes — framing delays as a measurable environmental liability with regulatory exposure for ESG stakeholders.
A dedicated deep-dive into 157K cancelled flights. Weather dominates at 66.28%, followed by carrier issues (23.95%), NAS (8.72%), and security (1.06%). 2022 saw the highest cancellation rate (6.18%), likely driven by post-COVID operational strain. The breakdown helps distinguish uncontrollable weather cancellations from carrier-driven failures where process improvements can reduce failure rates.
A structured five-stage pipeline from raw data to executive-ready dashboards.
Reviewed raw flight-level data. Handled nulls, cancellations, and delay edge cases.
Designed a star schema with dimension tables for airports, airlines, time, calendar, and routes. Managed active and inactive relationships for origin/destination logic.
Calculated delay metrics, cancellation rates, and reliability scores. Built route identifiers and performance flags.
Advanced measures using CALCULATE, RANKX, FILTER, and USERELATIONSHIP. Performance-aware aggregations.
Business-aligned dashboard design with clear KPI framing and benchmarking logic.
This project supports operational decision-making, identifies structural inefficiencies, and enables performance benchmarking for airline operations and network planning teams, airport authorities, aviation regulators, business and data analysts, and sustainability and ESG stakeholders.
Designing scalable data models for large datasets. Translating operational metrics into business KPIs. Using DAX for ranking, benchmarking, and conditional logic. Balancing analytical depth with executive-level clarity. Applying analytics beyond finance into operations and sustainability.
Predictive analysis implementation.
Open the .pbix file to explore interactive dashboards, or review the exported PDF for a static executive summary.
Operational view: Includes all positive delay minutes (ArrDelay > 0) for root cause and customer experience analysis.
Regulatory view: Includes only flights with ArrDelay ≥ 15 for BTS/DOT compliance and industry benchmarking (Source: Bureau of Transportation Statistics, Airline Service Quality Performance 234).
Data Source: Bureau of Transportation Statistics — On-Time Performance