memorandum

Subject: Predicting Flight Arrival Delay from Departure Delay

At our meeting on November 16th, while we were discussing the preliminary analysis of the arrival and departure data, you asked if we might be able to use that data to predict arrival times based on departure times. We reanalyzed the data for this purpose and want to share the results.

Don't use plagiarized sources. Get Your Custom Essay on
memorandum
Just from $13/Page
Order Essay

In this data set a change in departure delay has a high positive correlation to a change in arrival delay, 0.90 on a scale of -1.0 to +1.0. While the two are closely related, this alone does not prove causality or provide a measurement of predictability. For this, we conducted a statistical test called a regression analysis of the departure and arrival delays of the 444 random flights—169 United and 275 American—from LAX to JFK over a one year period. The analysis creates a formula from the data that best expresses the dependence of any arrival time based on that airplane’s departure time. We can use this formula to predict arrival times. The formula has two parts: 1) the arrival time in minutes saved (or lost) when the airplane departs on time, and 2) the weight or impact that each minute of departure delay has on arrival delay.

The analysis of this limited data set reveals that: 1) when airplanes depart on time they are likely to arrive about four minutes (4:07 minutes) earlier than scheduled, and 2) for every minute of departure delay there is approximately one minute (0:57 seconds) of arrival delay. For any given departure delay, say, 20 minutes, we can predict the arrival delay as: 4:07 minutes of early arrival + the weight of 0.96 times the 20 minute departure delay or approximately 15 minutes of arrival delay. The departure-arrival relationship of each of the 444 flights is depicted in the chart on the next page. This scatterplot chart (Figure 1) shows the linear relationship by drawing an ideal line through the data with the least amount of distance from all of the data points. The line through the dots representing individual airplane data is called the regression line and it is drawn by applying a formula to all of the data that mathematically minimizes the distance or variation.

The regression statistics also provide an estimate of the accuracy of the formula using two measures. The first measure is an indication of how much of the arrival delay is explained by the departure delay. In this data that measure is 82%, which means that 18% of arrival delays are due to factors other than departure delays. The other measure of accuracy is the range, and we can expect to see a range of 33 minutes (plus or minus 17 minutes of arrival delay around the regression line) for any given departure delay. This variation due to “other” factors is considerably less than if we were to make an arrival time estimate without using regression analysis.

American and United: Predicting Arrival Delay

y = 0.9558x – 4.1144


Multiple R = 0.9

R² = 0.8164

Std Error = 16.7

P value = 0.002288

Finally, the regression statistics include a measure of reliability. Our data approaches 100% reliability; meaning, if we were to collect the sample data again, we would have the same regression statistics.

While we need data on more flight delays to and from more airports, the early indication is that we will be able to predict arrival delays based on departure delays with considerable accuracy and reliability. We look forward to receiving a more comprehensive data set on which we could repeat this analysis.

If you have any questions or comments please contact us.

Calculate your order
Pages (275 words)
Standard price: $0.00
Client Reviews
4.9
Sitejabber
4.6
Trustpilot
4.8
Our Guarantees
100% Confidentiality
Information about customers is confidential and never disclosed to third parties.
Original Writing
We complete all papers from scratch. You can get a plagiarism report.
Timely Delivery
No missed deadlines – 97% of assignments are completed in time.
Money Back
If you're confident that a writer didn't follow your order details, ask for a refund.

Calculate the price of your order

You will get a personal manager and a discount.
We'll send you the first draft for approval by at
Total price:
$0.00
Power up Your Academic Success with the
Team of Professionals. We’ve Got Your Back.
Power up Your Study Success with Experts We’ve Got Your Back.
Live Chat+1(978) 822-0999EmailWhatsApp

Order your essay today and save 20% with the discount code GOODESSAY