Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Sin airport 2018 2020

•
•
Mohammed Awad
Mohammed Awad

Mohammed Salem Awadh provides forecasts for passenger traffic at Changi International Airport for 2018-2020. For annual forecasting, two trend models are used to minimize data discrepancies and set accurate targets. For monthly forecasting, seasonality patterns are determined without constraints. The document discusses two scenarios for each year, recommending the scenario with the lower predicted growth to avoid overestimating passenger levels. Overall, the forecasts provide a way to set targets and measure performance based on past traffic data and trends at Changi Airport.

1 of 5
Download to read offline
Changi International Airport Traffic Passengers Forecast (2018-2020) Creativity Cannot be born without Suffering And Success Cannot be achieved without Pains Dears, Forecasting is very important in our life, without the right expectation, plans cannot be setup, in many fields especially finance. (Preparing budgets), and performance can be developed if there is no targets, and accordingly we cannot know the right levels of KPIs, to measure the progress for what we are achieved . Today I am looking back for my previous work to see what progress or change that we made, off course numbers don’t lie, and that is the power of the knowledge, you can compare the forecasted results with the actual figure, to see the level of expectation – by defining the value of coefficient of determination ( R – Square). Good Case Study is Changi International Airport for 2018 – 2020 (period) Hope to enjoy Mohammed Salem Awadh Working In A Shadow 10 July 2020 By: Mohammed Salem Awad Aviation Consultant
Setting Target For Short Term Forecasting Introduction In short term forecasting, two objectives we address, First: Annual Forecasting – which defining the trend (Positive or Negative). Second: Monthly Forecasting – which defining the seasonality (Optimum Solution without constrains) First : Annual Forecasting – which defining the trend (Positive or Negative). The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2019. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2019 > Dec 2018. However,the outcomes can be represents by three directions 1- Positive Trends – both models in Upwards direction. The selection will be either the mid-point of positive trends Dec 2019 (Red Column) or the value of optimum solution (blue column) – select the one who has lower level, to avoid the high-risk passenger growth that will affect the final result. 2- Negative Trends – both models in downwards direction – Never set target with negative trend,here the best scenario is to look to Dec 2018 (Red Column) and the annual actual value for 2018 - select the one that has higher level. 3- High Discrepancy Data – and it is occur when we have one positive, and the other negative trend. i.e mean large gap between the two models, we follow step two provided that the mid-point Dec 2019 or optimum solution scenario are not higher than step two. In case they are higher, select the one that haslowerlevel. As in step one. Second: Monthly Forecasting – which defining the seasonality (Optimum Solution without constrains) This the main core program to define the seasonality pattern without any preset constrains, it follows the trends based on 36 months database. Which always comes closely to the preset values. Moreover, the blue column in the annual graph represents it. Therefore, we get a complete picture to select the right target. However, when it’s a higher value, then we have to select the preset values as a target. By: Mohammed Salem Awad Aviation Consultant Data Source: https://ec.europa.eu/eurostat/data/database
Airport Forecasting - 2020 (Issue No. 37) Changi Airport Singapore (SIN) Annual Passengers Forecast: The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2020. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2020 > Dec 2019. Scenario 1: Preset Annual Target = Passengers = 70,799,274 Pax. At 4.20 % Annual Growth (recommend) Scenario 2: Optimum Solution. = Passengers = 71,342,742 Pax. The objective is to minimize the risk of not achieving the desired goals. Both scenarios are fairs, and the results are good. The first scenario is fair enough to select,as it has lower growth, we recommend the 1st scenario, to avoid the high-risk passenger growth that mislead the final results. By: Mohammed Salem Awad Aviation Consultant Data Source: https://www.changiairport.com/corporate/our- expertise/air-hub/traffic-statistics.html
Airport Forecasting - 2019 Issue No. ( 10 / 2019) Airport Traffic Passengers Forecast - 2019 (SIN) The Dilemma ofperfect solutions As a general role, the model is fair when R-square is greater than 80% and S.T is less than ± 4, for our case these factors are very closely. Usually, we have two approaches: 1- Scenario 1 : Preset AnnualTarget = ForecastPax = 69,586,660 2- Scenario 2 : Optimum Solution. Forecast Pax = 69,411,794 (recommended) The objective is to minimize the risk of not achieving the desired goals. (Select, one who has lower growth). Both Scenarios are excellent and seasonality graphs are almost identical but the first one raise the target figure by 174,866 Pax. Therefore, we recommend the second one, to avoid the high- risk passenger growth. By: Mohammed Salem Awad Aviation Consultant Data Source: https://www.changiairport.com/corporate/about- us/traffic-statistics.html
Airport Forecasting - 2018 (Issue No. 106) Changi International Airport (SIN) Annually Forecast: The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. Here we implement two trend models by using Add a trend line in XLS sheet: First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2018. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2018 > Dec 2017. Monthly Forecast – we define the monthly targets that fulfill the condition of the first point (annual traffic setting forecast 2018 = 65,529,874 Pax) with minimum errors. Which shows a fair result in the second graph, at R-squared = 94.94 %, AnnualGrowth = 5.72 % and a clear picture about the seasonality pattern of the airport is defined as shown in table and graph. By: Mohammed Salem Awad Aviation Consultant Data Source:www.anna.aero

More Related Content

Sin airport 2018 2020

  • 1. Changi International Airport Traffic Passengers Forecast (2018-2020) Creativity Cannot be born without Suffering And Success Cannot be achieved without Pains Dears, Forecasting is very important in our life, without the right expectation, plans cannot be setup, in many fields especially finance. (Preparing budgets), and performance can be developed if there is no targets, and accordingly we cannot know the right levels of KPIs, to measure the progress for what we are achieved . Today I am looking back for my previous work to see what progress or change that we made, off course numbers don’t lie, and that is the power of the knowledge, you can compare the forecasted results with the actual figure, to see the level of expectation – by defining the value of coefficient of determination ( R – Square). Good Case Study is Changi International Airport for 2018 – 2020 (period) Hope to enjoy Mohammed Salem Awadh Working In A Shadow 10 July 2020 By: Mohammed Salem Awad Aviation Consultant
  • 2. Setting Target For Short Term Forecasting Introduction In short term forecasting, two objectives we address, First: Annual Forecasting – which defining the trend (Positive or Negative). Second: Monthly Forecasting – which defining the seasonality (Optimum Solution without constrains) First : Annual Forecasting – which defining the trend (Positive or Negative). The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2019. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2019 > Dec 2018. However,the outcomes can be represents by three directions 1- Positive Trends – both models in Upwards direction. The selection will be either the mid-point of positive trends Dec 2019 (Red Column) or the value of optimum solution (blue column) – select the one who has lower level, to avoid the high-risk passenger growth that will affect the final result. 2- Negative Trends – both models in downwards direction – Never set target with negative trend,here the best scenario is to look to Dec 2018 (Red Column) and the annual actual value for 2018 - select the one that has higher level. 3- High Discrepancy Data – and it is occur when we have one positive, and the other negative trend. i.e mean large gap between the two models, we follow step two provided that the mid-point Dec 2019 or optimum solution scenario are not higher than step two. In case they are higher, select the one that haslowerlevel. As in step one. Second: Monthly Forecasting – which defining the seasonality (Optimum Solution without constrains) This the main core program to define the seasonality pattern without any preset constrains, it follows the trends based on 36 months database. Which always comes closely to the preset values. Moreover, the blue column in the annual graph represents it. Therefore, we get a complete picture to select the right target. However, when it’s a higher value, then we have to select the preset values as a target. By: Mohammed Salem Awad Aviation Consultant Data Source: https://ec.europa.eu/eurostat/data/database
  • 3. Airport Forecasting - 2020 (Issue No. 37) Changi Airport Singapore (SIN) Annual Passengers Forecast: The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2020. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2020 > Dec 2019. Scenario 1: Preset Annual Target = Passengers = 70,799,274 Pax. At 4.20 % Annual Growth (recommend) Scenario 2: Optimum Solution. = Passengers = 71,342,742 Pax. The objective is to minimize the risk of not achieving the desired goals. Both scenarios are fairs, and the results are good. The first scenario is fair enough to select,as it has lower growth, we recommend the 1st scenario, to avoid the high-risk passenger growth that mislead the final results. By: Mohammed Salem Awad Aviation Consultant Data Source: https://www.changiairport.com/corporate/our- expertise/air-hub/traffic-statistics.html
  • 4. Airport Forecasting - 2019 Issue No. ( 10 / 2019) Airport Traffic Passengers Forecast - 2019 (SIN) The Dilemma ofperfect solutions As a general role, the model is fair when R-square is greater than 80% and S.T is less than ± 4, for our case these factors are very closely. Usually, we have two approaches: 1- Scenario 1 : Preset AnnualTarget = ForecastPax = 69,586,660 2- Scenario 2 : Optimum Solution. Forecast Pax = 69,411,794 (recommended) The objective is to minimize the risk of not achieving the desired goals. (Select, one who has lower growth). Both Scenarios are excellent and seasonality graphs are almost identical but the first one raise the target figure by 174,866 Pax. Therefore, we recommend the second one, to avoid the high- risk passenger growth. By: Mohammed Salem Awad Aviation Consultant Data Source: https://www.changiairport.com/corporate/about- us/traffic-statistics.html
  • 5. Airport Forecasting - 2018 (Issue No. 106) Changi International Airport (SIN) Annually Forecast: The best way to set up annual target and minimize the data discrepancy is to address the data by two trend models using the concept of 12 rolling months. Here we implement two trend models by using Add a trend line in XLS sheet: First – General Trend Model using the concept of Straight Line equation – defining general trend. Second – Most Recent Data Trend Model Using a Polynomial Model – Second-degree equation. This reflects the impact of most recent data on the path of general trend. The mid-point is the most convenient forecast annual result at Dec 2018. So as long as the gap between two models is small, the more accurate approaching value for setting annual target otherwise we have to select the half way distance between two extreme targets of these two models provided that Dec 2018 > Dec 2017. Monthly Forecast – we define the monthly targets that fulfill the condition of the first point (annual traffic setting forecast 2018 = 65,529,874 Pax) with minimum errors. Which shows a fair result in the second graph, at R-squared = 94.94 %, AnnualGrowth = 5.72 % and a clear picture about the seasonality pattern of the airport is defined as shown in table and graph. By: Mohammed Salem Awad Aviation Consultant Data Source:www.anna.aero