BTS Flight On-Time Performance Scraper

Overview

This Actor downloads and parses the Bureau of Transportation Statistics (BTS) On-Time Reporting Carrier flight performance dataset — the official US government source for airline delay, cancellation, and performance data. Access monthly flight data for all US domestic carriers with powerful filtering options.

Key Features:

• Free, no authentication required — Download directly from official BTS servers
• Monthly datasets — Full coverage from 1987 to present
• 600K+ flights per month — Complete US domestic flight records
• Rich filtering — By airline, airport, state, route, delay, cancellation status
• Computed fields — Delay categories, cancellation reasons, route keys for analysis
• Auto-updated — BTS publishes new data monthly

Data fields: Flight date, airline, flight number, tail number, origin/destination airports and cities, departure/arrival times and delays, cancellation/diversion status, distance, computed delay reasons (carrier, weather, NAS, security, late aircraft), and more.

Use Cases

• Airline Performance Analysis — Benchmark carriers by on-time performance, delay patterns, cancellation rates
• Route Planning & Optimization — Identify problematic routes, seasonal delay trends
• Travel Insurance Pricing — Historical delay data for risk modeling
• ML/AI Training Datasets — Clean, normalized flight data for delay prediction models
• Frequent Flyer Programs — Analyze airline reliability for loyalty decisions
• Airport Operations — Understand origin/destination impact on delays
• Research & Economics — Academic studies on transportation, supply chain disruption

Input Parameters

Parameter	Type	Default	Description
year	Integer	2024	Year for data (2010–2025)
month	Integer	12	Month (1–12)
originAirport	String	(all)	3-letter origin airport code (e.g., JFK, LAX, ORD)
destAirport	String	(all)	3-letter destination airport code (e.g., MIA, SFO)
carrier	String	(all)	Airline code (e.g., AA, DL, UA, WN, B6, AS)
cancelledOnly	Boolean	false	Only return cancelled flights
divertedOnly	Boolean	false	Only return diverted flights
delayedOnly	Boolean	false	Only return flights with ≥15 min departure delay
originState	String	(all)	2-letter state code for origin (e.g., CA, NY, TX)
destState	String	(all)	2-letter state code for destination (e.g., FL, HI)
maxResults	Integer	5000	Max records to return (0 = unlimited; monthly datasets contain 600K+ rows)

Output Fields

Each record includes:

Flight Identifiers:

• flightDate, year, month, dayOfMonth, dayOfWeek
• carrier (airline code), flightNumber, tailNumber

Route Information:

• origin, originCity, originState, originStateName
• dest, destCity, destState, destStateName
• routeKey (computed: origin-dest, e.g., JFK-LAX)
• distance (miles)

Departure:

• crsDepTime (scheduled), depTime (actual), depDelay (signed minutes)
• depDelayMinutes (absolute minutes late; null if on-time or early)
• depDel15 (boolean: delayed ≥15 minutes)
• taxiOut, wheelsOff

Arrival:

• crsArrTime (scheduled), arrTime (actual), arrDelay (signed minutes)
• arrDelayMinutes, arrDel15 (boolean: delayed ≥15 minutes)
• wheelsOn, taxiIn

Elapsed Time:

• crsElapsedTime, actualElapsedTime, airTime (in minutes)

Cancellation & Diversions:

• cancelled (boolean)
• cancellationCode (A=Carrier, B=Weather, C=NAS, D=Security)
• cancellationReason (mapped: "Carrier", "Weather", "NAS", "Security")
• diverted (boolean)

Delay Breakdown (sum to total departure delay):

• carrierDelay (minutes — airline operational issue)
• weatherDelay (minutes — weather at origin/destination)
• nasDelay (minutes — NAS congestion)
• securityDelay (minutes — security issue)
• lateAircraftDelay (minutes — previous flight delay)

Computed Fields:

• delayCategory (one of: on-time, delayed-minor [15–60 min], delayed-major [>60 min], cancelled, diverted)

Tutorial: Analyze JFK→LAX Delays

1. Open the Actor Input Form in Apify Console
2. Configure inputs:
   • year: 2024
   • month: 12
   • originAirport: JFK
   • destAirport: LAX
   • maxResults: 500
3. Click Run
4. Wait for completion (30–90 seconds depending on filters; larger datasets take longer)
5. View Results:
   • Open the Dataset tab
   • Analyze depDelayMinutes, arrDelayMinutes, delayCategory
   • Sort by delay to identify worst-performing flights
   • Filter by carrier to compare airlines on this route

Example insights:

• "JFK→LAX has avg departure delay of 12 minutes in December"
• "Southwest (WN) had 3 cancellations, United (UA) had 0"
• "Afternoon flights (14:00–18:00) show 18% more delays than morning"

Tutorial: Find Airlines with Most Cancellations

1. Inputs:
   • year: 2024
   • month: 12
   • cancelledOnly: true
   • maxResults: 0 (unlimited)
2. Run → captures all cancellations for December 2024
3. Export as CSV/JSON
4. Analyze in spreadsheet or Python:

   df = read_csv('dataset.csv')
   cancellations_by_carrier = df.groupby('carrier').size().sort_values(ascending=False)
   print(cancellations_by_carrier)
   

Data Schema & Completeness

Source: Official Bureau of Transportation Statistics (BTS), U.S. Department of Transportation

Coverage: All US-based carriers operating domestic flights

Data freshness: BTS publishes preliminary data within 1–2 days of month end; final data within 2–3 weeks. This Actor downloads the latest available dataset for the requested month/year.

Completeness: Fields are populated per BTS schema. Delays and cancellation reasons may be null if the flight was:

• Cancelled before departure (no delay values)
• Diverted (alternate dataset)
• Early arrival (negative depDelay)

Note on large datasets: December and summer months (June–August) have 600K+ rows. Use filters (originAirport, carrier, delayedOnly, maxResults) to keep compute costs manageable.

Common Filters & Examples

Use Case	Input Config
All Southwest flights in April 2025	carrier: "WN", year: 2025, month: 4
Cancelled flights in California	originState: "CA", cancelledOnly: true
JFK departures in Dec 2024, first 100	originAirport: "JFK", maxResults: 100
Morning flights (05:00–12:00)	No built-in filter; post-process crsDepTime
Major delays on top 10 routes	delayedOnly: true, maxResults: 50000 → group by routeKey

Pricing

• Per-result: $0.003/record
• Compute units: 1 run ≈ 0.5–2 compute units depending on dataset size and filters
• Example: 5,000 records = ~$0.015 + 0.5 compute units ($0.025) = ~$0.04 total
• Full monthly dataset (600K rows, no filters) ≈ $1.80 results + ~$1–2 compute = $4–5 total

Cost optimization:

• Use maxResults to limit output
• Filter by carrier or originAirport to reduce processing
• Export monthly, not weekly

Troubleshooting

Q: "Failed to download BTS data" error

• Check that year is 2010–2025 and month is 1–12
• Older data (pre-2010) is not available via this URL pattern
• BTS may delay monthly publishing by 1–2 weeks; try the prior month

Q: Got 0 results with filters applied

• Your filter combination may have no matches (e.g., carrier "FX" doesn't exist)
• Check airport codes are valid 3-letter IATA codes
• State codes must be 2-letter (e.g., "CA" not "California")

Q: Dataset very large, run timed out

• Monthly datasets (600K+ rows) can take 2–5 minutes to parse
• Use maxResults to cap output
• Add delayedOnly: true or filter by carrier/originAirport

Q: Data doesn't match BTS website directly

• This Actor downloads the raw "Reporting Carrier" dataset from BTS
• BTS website may show aggregated or summarized data; raw CSVs are authoritative
• Dates are in ISO format (YYYY-MM-DD); BTS website may display differently

Similar Actors

• US Airport Operations Tracker — Real-time airport delays and statistics (complement to historical BTS data)
• Flight Price Tracker — Current flight prices and availability
• Airline News Monitor — Track airline announcements and alerts

Legal & Attribution

Open Data: BTS flight data is public domain. No authentication or payment to BTS is required.

Proper Attribution: When publishing analysis:

• Cite "U.S. Department of Transportation, Bureau of Transportation Statistics"
• Include: "Data source: BTS On-Time Reporting Carrier dataset"

Disclaimer: This Actor provides data as-is. Apify and the creator are not liable for inaccuracies or delays in BTS data publishing. For regulatory/legal decisions, refer to official BTS sources directly.

Terms: Use this Actor only for lawful purposes. Reselling BTS data without attribution violates open data principles. Proper attribution is requested but not legally required (BTS is public domain).

Support

• Questions? Check this README or contact support via Apify
• Data quality issues? Report directly to BTS — this Actor mirrors their official datasets
• Actor bugs? File an issue or contact the developer

---

Last updated: May 2025
Version: 0.1