ASSIGNMENT # 1 – FORECASTING
Attached is traffic data for Lester B. Pearson International Airport for the years 1988 through 2012. From this data:
1. ( a ) Prepare and discuss forecasting models for total E&D Passengers using a simple linear relationship as follows:
(i) As a function of yield alone ( i.e.. P = a(Yield) + b where P = Annual E&D Passengers).
(ii) As a function of GDP alone ( i.e. P = a(GDP) + b where P = Annual E&D Passengers).
(iii) As a function of both yield and GDP (i.e. P = a(GDP) + b(Yield) + c where P = E&D Passengers).
Include all the relevant tests for significance and model validity for these simple models. Discuss the model results. Is the overall model result (i.e. adjusted R2) significant ? Are the explanatory variables statistically significant? Do the coefficient signs make sense? What about the residuals? Which appears to be the best model?
( b ) Is the model fit for model a(ii), the GDP model, improved by using a lagging relationship (i.e. lagging the GDP variable one year so that for example the GDP in 1989 explains the traffic in 1990 etc.)? Why might you expect that the model fit wouldt be improved by using a lagging relationship (i.e. how does this fit with possible traveler behavior)?
( c ) Presumably traffic to the United States is influenced both by the state of the Canadian economy and the state of the economy of the United States. Develop a forecasting model for the trans-border traffic sector alone by using an index which equally weights the GDPs of the USA and Canada as the explanatory variable? Hint: Find the real dollar values of the US GDP from the web for the years 1988 to 2012 and develop an index set at 50 for 1988 using the year over year changes in real USGDP values. Develop a similar index for the Canadian GDP data. Produce a combined (i.e. set at 100 for 1988) index weighting the Canadian and US indexes equally by adding these two values. Use this combined index in place of the Canadian GDP in model a(ii) and carry out the regression using the trans-border traffic figures only (i.e. P(TB) = a(GDP Index) + b where Index = for year “n” equals 50 x real change in Cdn GDP to year “n” plus 50 times real change in USA GDP to year “n” and where P(TB) is the trans-border traffic. How does model compare with simply using the total traffic model developed in (a) (ii) above and apportioning the traffic by market share? Can you suggest a similar approach to develop a sector specific model to forecast the international tarffic segment (hint: what might be a comparable index)?
Note: for parts (b) and (c) only the adjusted “R” square needs to reported to support your answer, you do not have to include all the tests for significance as you did in part (a).
2) Answer the following three questions regarding traffic forecast modeling for Pearson.:
(a) Porter Airlines startup from the Island Airport (Billy Bishop Airport) is siphoning traffic away from Toronto Pearson International Airport. Recalibrate model a (ii) adding in the Porter traffic data to the Pearson total traffic to make a true Greater Toronto Area (GTA) model. Is this model for the GTA a better fit than the model for Pearson alone based on the adjusted R2?
(b) Traffic is also being siphoned away by the low cost carriers operating out of Buffalo Airport. It is suspected that the Toronto traffic being diverted to Buffalo is somewhere between 750,000 to 1,500,0000 passengers per year but there is no way of knowing for sure. This loss began in the early 2000’s when the US low cost carriers set up operations in Buffalo. Buffalo airport does not have (or at least does not chose to disclose) any data regarding the number of canadians using their facility. Given these facts can you suggest anyway this issue can be given consideration in formulating the Pearson traffic forecasting model?
(c) Pearson is a major international hub airport and as such approximately one quarter of the traffic is connecting – that is passengers are arriving from one airport and then changing planes to continue their journey to another airport. While this connecting traffic share has been relatively constant up to 2005 the airports strategy since 2005 has been to grow this connecting activity aggressively by actively encouraging carriers to connect traffic at Toronto as shown in the data. How might this connecting traffic share be impacting the forecast? How could you improve the forecast model knowing this fact?
3) Using model a(ii) from question #1 above prepare forecasts of future E&D Passenger traffic for L.B. Pearson International Airport for the years 2017, 2022 and 2027 based on the following assumptions concerning the GNP explanatory variable (all figures are in annual % growth)
2013 – 2017 2018 – 2022 2023 – 27
Optimistic (Real GNP) +2.5% +3% +2.0%
Expected (Real GNP) +1.0% +2.0% +1.0%
Pessimistic (Real GNP) -0.5% +1.0% +0.5%
For each forecast calculate the expected propensity to travel (annual trips / capita) by air by dividing by the expected population over the period of the forecast . How reasonable do the numbers appear for each of the three forecasts in the year 2027 ?
Note: I have not provided you with the GTA population historical data nor any projections. I want you to find this information yourself to give you some idea of the effort you need to make to research the inputs to the model.
4) For the forecast in the “Expected Case” only in 3) above estimate the associated air carrier itinerant movements based on the following facts and assumptions:
i) in 2012 their were 433,900 air carrier itinerant movements and the average involvement ratio
ii) average aircraft size is expected to increase by one seat per year over the entire forecast
iii) the involvement ratio is expected to increase by 0.05% per year over the entire forecast
5) Assume a 90th percentile planning criteria. That is to say assume we will have as a policy the intent to serve demand such that for 10 % of the days the level of service will be less than the target level. Based on this 90th per centile criteria it is estimated that the ratio of planning day to annual movement traffic
is 1 / 285 .
( a ) Using this 90th percentile planning day criteria forecast the daily movements for the planning day in the years 2017, 2022 and 2027.
( b ) Estimate the hourly movements for 2017 if the same daily movement pattern (distribution) is assumed to exist for the planning day as existed on Thursday, 12 August, 2004 ( note: the 12 August 2004 hourly profile is presented in the data set ). What is the peak hour traffic volume?
( c ) Estimate the peak hour movements for 2022 and 2027 if the same daily movement pattern (distribution) is assumed to exist for the planning day as existed on Thursday, 12 August, 2004 except the peak hour is flattened by ½ of 1% per year starting in year 2018 (i.e. after 5 years the peak share of daily traffic would be 2.5% less than it was in 2012).
( d ) If the average gate occupancy time (i.e. turn around time) of the aircraft using Pearson is found to be 50 minutes approximately how many gates will be required to serve the peak hour traffic in 2017, 2022 and 2027?
( e ) If the hourly capacity of the current runway system is estimated to be 110 movements per hour in what year would you predict the airport airside will reach capacity during the peak hour. If the ultimate capacity of the runway system when fully expanded is 136 movements per hour when will the entire airport reach full capacity?
6) Using the data for question #6 provided from the Official Airline Guide:
(a) Prepare a linked nominal schedule from the OAG data provided for a five gate airport terminal.
(b) From the linked nominal schedule prepare a gate assignment plot for the airport terminal building. Note: three of the gates are domestic gates, one is a dedicated international gate and one is a swing gate (i.e. can either be a domestic or international gate) . Apron parking space is available to store two aircraft awaiting for gates to become available. Note that aircraft of less than 99 seats require a five (5) minute buffer, those of greater size require a ten (10) minute buffer.
(c) From the nominal schedule prepare a profile of the number of passengers in the departure process at the airport during the period of the Nominal Schedule using the distribution of arrival times for both domestic and international flights.
Delivering a high-quality product at a reasonable price is not enough anymore.
That’s why we have developed 5 beneficial guarantees that will make your experience with our service enjoyable, easy, and safe.
You have to be 100% sure of the quality of your product to give a money-back guarantee. This describes us perfectly. Make sure that this guarantee is totally transparent.Read more
Each paper is composed from scratch, according to your instructions. It is then checked by our plagiarism-detection software. There is no gap where plagiarism could squeeze in.Read more
Thanks to our free revisions, there is no way for you to be unsatisfied. We will work on your paper until you are completely happy with the result.Read more
Your email is safe, as we store it according to international data protection rules. Your bank details are secure, as we use only reliable payment systems.Read more
By sending us your money, you buy the service we provide. Check out our terms and conditions if you prefer business talks to be laid out in official language.Read more