This document outlines an assignment for a business data analysis unit. Students are asked to complete individual reports analyzing data from Southwest Airlines. The assignment involves conducting exploratory data analysis on passenger booking and show-up data, using probability distributions and Monte Carlo simulation to model the airline's overbooking strategy. Students must simulate different booking levels, compute flight profits, and recommend an overbooking strategy. They are also asked to explain how the simulation could evaluate other booking levels and inform overbooking policies. The report must apply techniques of data analysis, modeling, and simulation to evaluate statistical evidence and communicate findings clearly in a technical report.
This document outlines an assignment for a business data analysis unit. Students are asked to complete individual reports analyzing data from Southwest Airlines. The assignment involves conducting exploratory data analysis on passenger booking and show-up data, using probability distributions and Monte Carlo simulation to model the airline's overbooking strategy. Students must simulate different booking levels, compute flight profits, and recommend an overbooking strategy. They are also asked to explain how the simulation could evaluate other booking levels and inform overbooking policies. The report must apply techniques of data analysis, modeling, and simulation to evaluate statistical evidence and communicate findings clearly in a technical report.
This document outlines an assignment for a business data analysis unit. Students are asked to complete individual reports analyzing data from Southwest Airlines. The assignment involves conducting exploratory data analysis on passenger booking and show-up data, using probability distributions and Monte Carlo simulation to model the airline's overbooking strategy. Students must simulate different booking levels, compute flight profits, and recommend an overbooking strategy. They are also asked to explain how the simulation could evaluate other booking levels and inform overbooking policies. The report must apply techniques of data analysis, modeling, and simulation to evaluate statistical evidence and communicate findings clearly in a technical report.
This document outlines an assignment for a business data analysis unit. Students are asked to complete individual reports analyzing data from Southwest Airlines. The assignment involves conducting exploratory data analysis on passenger booking and show-up data, using probability distributions and Monte Carlo simulation to model the airline's overbooking strategy. Students must simulate different booking levels, compute flight profits, and recommend an overbooking strategy. They are also asked to explain how the simulation could evaluate other booking levels and inform overbooking policies. The report must apply techniques of data analysis, modeling, and simulation to evaluate statistical evidence and communicate findings clearly in a technical report.
Download as DOCX, PDF, TXT or read online from Scribd
Download as docx, pdf, or txt
You are on page 1of 6
1
0Unit Title: Business Data Analysis
Unit Code: BSS002-6 Assessment 1: Individual Report (50%) It is expected that each student work independently on all the tasks set out in this assignment, applying problem solving strategies and selecting appropriate techniques in modelling and simulation in order to produce an individual report according to the guidelines provided. This individual work must be your own work only with no consultation with others. The individual report must not exceed 2,000 words. There will be an oral viva ONLY for those individual reports which exceed the similarity index of 25% produced by Turnitin. Assignment Title: Southwest Airlines Issued date: Week One: Friday 16/05/2014 Submission due date: Week Nine: Friday 11/07/2014: A single report per student should be submitted via the TURNITIN link in the Assessments folder and a paper copy should also be submitted to the Assignment Collection Centre. Unit co-ordinators name: C F Shoostarian Learning outcomes: Core Learning Outcomes
Learning Outcomes On completion of this unit you should be able to: Assessment Criteria To achieve the learning outcome you must demonstrate the ability to: 1 Conduct an Exploratory Data Analysis to critically appraise the data Construct appropriate charts and statistical summary measures according to the nature of the data
Assess the hidden features of the data and its distribution
Critically appraise the discrepancy between the empirical distributions of model and data 2 Formulate appropriate statistical models of random events Select an appropriate probability distribution model based on: the nature of the data; the Exploratory Data Analysis conducted; testing assumptions. 2
3 Plan, design, formulate and construct business models where there is uncertainty present Use Monte Carlo Simulation to develop models of situations where there is uncertainty in the problem such as: demand and use what if analysis to consider possible alternatives 4 Evaluate statistical evidence Assess the evidence provided by the data in favour of some claim about the population using: Confidence interval Scientific method of investigation 5 Research, synthesise and produce a technical report Work effectively both in group settings and independently to communicate your findings clearly and concisely. Recommend solutions and provide a better insight to the problem, justifying your choice in a technical report which has: clear structure and logical sequence; tables, diagrams and figures are appropriate, clearly labelled and interpreted; accuracy; objectivity; continuity; conciseness; A relatively formal writing style should be used when composing technical reports. Assessment Criteria: Learning Outcomes with the corresponding Assessment Criteria Grading/ Criteria A+ A A- B+ B B- C+ C C- D+ D D- E 16 15 14 13 12 11 10 9 8 7 6 5 4 LO1: Conduct an Exploratory Data Analysis to critically appraise the data For Tasks 1 3: .Construct appropriate charts and statistical summary measures according to the nature of the data .Assess the hidden features of the data and its distribution .Critically appraise the discrepancy between the empirical distributions of model and data comprehensively For Tasks 1 3: .Construct appropriate charts and statistical summary measures according to the nature of the data .Assess the hidden features of the data and its distribution accurately
For Tasks 1 3: .Construct appropriate charts and statistical summary measures according to the nature of the data .Assess the hidden features of the data and its distribution with minor errors
For Tasks 1 3: .Construct appropriate charts and statistical summary measures according to the nature of the data .Assess the hidden features of the data and its distribution with major errors
For Tasks 1 3: .Construct appropriate charts and statistical summary measures according to the nature of the data
3
Grading/ Criteria A+ A A- B+ B B- C+ C C- D+ D D- E 16 15 14 13 12 11 10 9 8 7 6 5 4 LO2: Formulate appropriate statistical models of random events Tasks 1, 2 and 3: Select an appropriate probability distribution model based on: the nature of the data; the Exploratory Data Analysis conducted; testing assumptions. Tasks 1, 2 and 3: Select an appropriate probability distribution model based on: the nature of the data; the Exploratory Data Analysis conducted. Tasks 1, 2 and 3: Select an appropriate probability distribution model based on: the nature of the data; the Exploratory Data Analysis conducted; with some errors Tasks 1, 2 and 3: Attempting to develop models for the given data Tasks 1, 2 and 3: Exploring to develop models for the given data
Grading/ Criteria A+ A A- B+ B B- C+ C C- D+ D D- E 16 15 14 13 12 11 10 9 8 7 6 5 4 LO3: Plan, design, formulate and construct business models where there is uncertainty present Use Monte Carlo Simulation to develop models that will simulate the performance of the overbooking system and conduct a comprehensive what if analysis to consider possible alternatives Use Monte Carlo Simulation to develop models that will simulate the performance of the overbooking system and conduct what if analysis to consider possible alternatives Use Monte Carlo Simulation to develop models that will simulate the performance of the overbooking system and conduct what if analysis to consider possible alternatives with some errors Attempt at Using Monte Carlo Simulation to develop models that will simulate the performance of the overbooking system Explore the use of Monte Carlo Simulation to develop models that will simulate the performance of the overbooking system
Grading/ Criteria A+ A A- B+ B B- C+ C C- D+ D D- E 16 15 14 13 12 11 10 9 8 7 6 5 4 LO5: Collaboratively research, synthesize and produce a technical report for effective communication using a complex data source. Work effectively independently to communicate your findings clearly and concisely. Recommend solutions and provide a better insight to the problem, justifying your choice in a technical report which has: clear structure and logical sequence; tables, diagrams and figures are appropriate, clearly labeled and interpreted; accuracy; objectivity; continuity; conciseness; Work effectively independently to communicate your findings clearly and concisely. Recommend solutions and provide a better insight to the problem, justifying your choice in a technical report which has clear structure with appropriate diagrams and accurate discussions A structured business report. Calculations and discussions with some errors. A business report. Calculations and discussions with errors. A business report and tasks have been attempted.
4
Southwest Airlines More than 38 years ago, Rollin King and Herb Kelleher got together and decided to start a different kind of airline. They began with one simple notion: If you get your passengers to their destinations when they want to get there, on time, at the lowest possible fares, and make darn sure they have a good time doing it, people will fly your airline. And you know what? They were right. Southwest Airlines operates a commuter flight between Orlando City in the central region of the U.S. state of Florida and Long island which is an island in the U.S. state of New York. The plane holds 30 passengers, and the airline makes 100 profit on each passenger on the flight. When South Central takes 30 reservations for the flight, experience has shown that on average, two passengers do not show up. As a result, with 30 reservations, Southwest is averaging 28 passengers with a profit of 28(100) = 2,800 per flight. The airline operations office has asked for an evaluation of an overbooking strategy where they would accept 32 reservations even though the airplane holds only 30 passengers. The probability distribution for the number of passengers showing up when 32 reservations are accepted is as follows: Passengers showing up Probability 28 0.05 29 0.25 30 0.50 31 0.15 32 0.05 The airline will receive a profit of 100 for each passenger on the flight up to the capacity of 30 passengers. The airline will incur a cost of any passenger denied seating on the flight. This cost covers added expenses of rescheduling the passenger as well as loss of goodwill, estimated to be 150 per passenger. Task 1 (a) Develop a worksheet model that will simulate the performance of the overbooking system. Simulate the number of passengers showing up for each of 500. Use the results to compute the profit for each flight. 5
(b) Does your simulation recommend the overbooking strategy? What is the mean profit per flight if overbooking is implemented? (c) Explain how your simulation model could be used to evaluate other overbooking levels such as 31, 33, 34 and for recommending overbooking strategy. The management at Southwest Airlines is interested in using simulation to estimate the profit per unit for a new product they have developed called EarlyBird Check-In SM . The Probability distributions for the purchase cost, the labour cost and the transportation cost for this product are given in the table below: Purchase Cost () Probability Labour Cost () Probability Transportation Cost () Probability 10 0.25 20 0.10 3 0.75 11 0.45 22 0.25 5 0.25 12 0.30 24 0.35 25 0.30 Assuming that these are the only costs and that the selling price for the product will be 45 per unit. Task 2 (a) Provide the base-case, worst-case and best-case calculations for the profit per unit. (b) Set up intervals of random numbers that can be used to randomly generate the three cost components. Using the random numbers 0.1862, 0.7466 and 0.6171, calculate the profit per unit. (c) Management believes the project many not be profitable if the profit per unit is less than 5. Explain how simulation can be used to estimate the probability the profit per unit will be less than 5. Task 3 The management at Southwest Airlines is interested in using simulation to estimate the profit per unit for a new product they have developed called Android App. The Airlines advertising campaign for this new product is as follows:
Whether you're planning your next trip or checking in for your flight, the Southwest Airlines Android App makes travelling even more convenient. From the App you can check flight status, change reservations, add EarlyBird Check-In SM to an existing reservation, access your Rapid Rewards account and more while you're on the go! The fixed cost to begin the production of the product is 30,000. The variable cost for the product is uniformly distributed between 16 and 24 per unit. The product will sell for 50 per unit. Demand for the product is best described by a normal probability distribution with a mean of 1200 units and a standard deviation of 300 units. Use 500 simulation trials to answer the following questions. 6
(a) What is the mean profit for the simulation? (b) What is the probability the project will result in a loss? (c) What is your recommendation concerning the introduction of the product?
Data Forecasting and Segmentation Using Microsoft Excel: Perform data grouping, linear predictions, and time series machine learning statistics without using code