Airline On Time Performance Systems Design Project by Matthias Chan.

Slides:



Advertisements
Similar presentations
U.S. and foreign air carriers operating passenger flights in, to, or from the U.S. with at least one aircraft with an original passenger capacity of 30.
Advertisements

Forecasting Introduction
AGIFORS--RM Study Group New York City, March 2000 Lawrence R. Weatherford, PhD University of Wyoming Unconstraining Methods.
Chap 12-1 Statistics for Business and Economics, 6e © 2007 Pearson Education, Inc. Chapter 12 Simple Regression Statistics for Business and Economics 6.
Guide to Using Excel 2007 For Basic Statistical Applications To Accompany Business Statistics: A Decision Making Approach, 8th Ed. Chapter 16: Analyzing.
Types of Data This module was developed by Business Process Improvement. For more modules, please contact us at or visit our website
2 Excel* MegaStat Minitab SPSS JMP POM *We will focus on this readily available software in the demonstrations to follow Statistical Software.
Regression Analysis Once a linear relationship is defined, the independent variable can be used to forecast the dependent variable. Y ^ = bo + bX bo is.
Regression and Correlation
1 Spreadsheet Modeling & Decision Analysis: A Practical Introduction to Management Science, 3e by Cliff Ragsdale.
MOVING AVERAGES AND EXPONENTIAL SMOOTHING
Chapter 13 Forecasting.
SIMPLE LINEAR REGRESSION
McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. Time Series and Forecasting Chapter 16.
Time Series and Forecasting
Plane Spotting 2 Or how to get more free peanuts. Predicting Air Delays at SeaTac airport. A CS 773C Project by Sam Delaney.
Correlation and Linear Regression
Correlation and Linear Regression
McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. Chapter 13 Linear Regression and Correlation.
Correlation and Linear Regression Chapter 13 Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin.
SIMPLE LINEAR REGRESSION
Linear Regression and Correlation
Correlation and Linear Regression
Graphical Analysis. Why Graph Data? Graphical methods Require very little training Easy to use Massive amounts of data can be presented more readily Can.
CPE 619 Simple Linear Regression Models Aleksandar Milenković The LaCASA Laboratory Electrical and Computer Engineering Department The University of Alabama.
1 Least squares procedure Inference for least squares lines Simple Linear Regression.
CHAPTER 15 Simple Linear Regression and Correlation
CHAPTER 14 MULTIPLE REGRESSION
1 Spreadsheet Modeling & Decision Analysis: A Practical Introduction to Management Science, 3e by Cliff Ragsdale.
Chap 12-1 A Course In Business Statistics, 4th © 2006 Prentice-Hall, Inc. A Course In Business Statistics 4 th Edition Chapter 12 Introduction to Linear.
Linear Regression Least Squares Method: the Meaning of r 2.
Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
LMINET2: An Enhanced LMINET Dou Long, Shahab Hasan December 10, 2008.
McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. Chapter 13 Linear Regression and Correlation.
Time Series Analysis and Forecasting
Sullivan – Fundamentals of Statistics – 2 nd Edition – Chapter 4 Section 2 – Slide 1 of 20 Chapter 4 Section 2 Least-Squares Regression.
F E D E R A L A V I A T I O N A D M I N I S T R A T I O N A I R T R A F F I C O R G A N I Z A T I O N 1 Federal Aviation Administration System Airport.
5 Descriptive Statistics Chapter 5.
Chapter Thirteen Copyright © 2004 John Wiley & Sons, Inc. Sample Size Determination.
The McGraw-Hill Companies, Inc. 2006McGraw-Hill/Irwin DSS-ESTIMATING COSTS.
© The McGraw-Hill Companies, Inc., 2002 McGraw-Hill/Irwin Slide Cost Estimation.
Copyright © 2010 Pearson Education, Inc Chapter Seventeen Correlation and Regression.
Introduction to Regression
Simple Linear Regression The Coefficients of Correlation and Determination Two Quantitative Variables x variable – independent variable or explanatory.
Time Series and Forecasting Chapter 16 McGraw-Hill/Irwin Copyright © 2012 by The McGraw-Hill Companies, Inc. All rights reserved.
Simple linear regression and correlation Regression analysis is the process of constructing a mathematical model or function that can be used to predict.
Managerial Decision Modeling 6 th edition Cliff T. Ragsdale.
Chapter 15 Forecasting. Forecasting Methods n Forecasting methods can be classified as qualitative or quantitative. n Such methods are appropriate when.
Chapter 13 Linear Regression and Correlation. Our Objectives  Draw a scatter diagram.  Understand and interpret the terms dependent and independent.
Correlation and Linear Regression
Chapter 2 Linear regression.
Multiple Regression.
Chapter 13 Simple Linear Regression
Unit 4 LSRL.
Forecasting Methods Dr. T. T. Kachwala.
Statistical Quality Control, 7th Edition by Douglas C. Montgomery.
Correlation and Simple Linear Regression
Regression Analysis Part D Model Building
Mixed Costs Chapter 2: Managerial Accounting and Cost Concepts. In this chapter we explain how managers need to rely on different cost classifications.
Chapter 5 LSRL.
S519: Evaluation of Information Systems
Multiple Regression.
Chapter 5 LSRL.
Estimating the number of components with defects post-release that showed no defects in testing C. Stringfellow A. Andrews C. Wohlin H. Peterson Jeremy.
DSS-ESTIMATING COSTS Cost estimation is the process of estimating the relationship between costs and cost driver activities. We estimate costs for three.
Warm-up: Pg 197 #79-80 Get ready for homework questions
Forecasting Plays an important role in many industries
Correlation and Simple Linear Regression
Correlation and Simple Linear Regression
Presentation transcript:

Airline On Time Performance Systems Design Project by Matthias Chan

Agenda Introduction Definitions Procedural Details Model – Words Model – Equations Results Demo Analysis Conclusion References

Introduction There are over flight each day in the United States Being “On Time” is defined as being 15 minutes of a scheduled departure and/or arrival time About 70% of all flights are “On Time” Data from the last 21 years is available to the public (starting from 1988)

Definitions Total Delay is defined as the sum of the amount of delay from the origin airport and the amount of delay at the destination airport. Total Adjusted Delay is defined as the sum of the difference between actual and scheduled departure time and difference between actual and scheduled arrival time. Total Adjusted Delay differs from Total Delay in that Total Adjusted Delay takes early departures and early arrivals and assigns a negative value of delay for them.

Procedural Details The Bureau of Transportation Statistics (BTS) is a subdivision of the U.S. Department of Transportation TranStats is a department of the BTS and reports data for each flight in the domestic U.S Using TranStats, we were able to collect data from all the flights in the past 21 years

Procedural Details

Each month of every year ( ) were downloaded in csv format Each file was 30 MB and each month had, on average, 370,000 flights In total, the data we collected was about 9.2 GB’s worth of data

Procedural Details We then took each month, found the control limits of the data, and then created summary reports in excel From the summary reports, we observed different variables (i.e. Day of Week, Distance, etc.) and found that seven of the variables looked like they had an effect on delay (Year, month, day, day of week, airline, origin airport and destination airport)

Procedural Details We then found 25 random flights in each month, and create data sets of 6200 flights that we wanted to use in calculating estimated delay

Model - Words Minimize the square error of estimated delay from actual delay By varying the weights given to each of the seven parameters We made the decision not to add non negativity constraints because of the way delay was defined (negative delay means an early flight)

Model - Equations A i = Coefficient of delay for variable I Y j = Average delay of year j where j = 1988:2009 M k = Average delay of month k where k = 1:12 D l = Average delay of day l where l = 1:31 W n = Average delay of week n where n = 1:7 Air p = Average delay of airline p where p = 1:19 O r = Average delay of origin airport r where r = 1:279 De r = Average delay of destination airport where r = 1:279 Delay(d) = delay of flight d where d = 1:6200 A = [A 1 A 2 A 3 A 4 A 5 A 6 A 7 ]’ data = [Y year M month D day W week Air airline­ O origin De destination ]

Model - Equations A = [A 1 A 2 A 3 A 4 A 5 A 6 A 7 ]’ data = [Y year M month D day W week Air airline­ O origin De destination ]

Results A = [A 1 A 2 A 3 A 4 A 5 A 6 A 7 ]’ data = [Y year M month D day W week Air airline­ O origin De destination ] Notice that there is an emphasis on Day of Week and Origin TrialA1A2A3A4A5A6A7 Mean Error Average

Demo Matlab GUI – Shown Below

Analysis We took our model and tested its predictive strength against the newest available data Data for February 2009 was released recently, we tested our estimates to see how many flights were accurately predicted As we suspected, 70.84% of the flights in February were in the range that we predicted

Analysis We took a look at how 9/11 effected delay in the U.S We found that delay actually decreased around 9/11 and the few years after it This can be attributed to several factors, less optimism by the airlines, tighter security (keep in mind that this is scheduled delay, not security delays)

Analysis

Conclusion We discovered that there can be an effective way to predict flight delay given all the data of past years However, we can only predict the data based on flights that are “in control” (70%) In the future, the project could expand to include a nonlinear equation to model delay, include international flights as well, include types of delay (i.e. weather, carrier, etc.), predict coefficient of year in question, instead of just using previous year.

References “Transtats – BTS.” Montgomery, Douglas C. Introduction to Statistical Quality Control. 5 th Edition. John Wiley & Sons, Hillier, Frederick S. and Lieberman, Gerald J. Intoduction to Operations Research. 8 th Edition. McGraw-Hill Higher Education

Questions?