Restarting aviation: How proactive forecasting and planning will lead the way to recovery

Guest Author by Guest Author | May 12, 2021

Written by Rasmus Kaster, Director Consultancy, Copenhagen Optimization and Emilie Knutz, Consultant, Copenhagen Optimization

COVID-19 has presented airports with a range of new challenges. The ongoing COVID-19 pandemic has forced governments to re-impose restrictions as many states are still confronted to subsequent waves of infections. The pace at which those restrictions will be able to be lifted will dictate how soon traffic recovery will commence. In this specific context, historical data can no longer be relied on for traffic scheduling and operational planning.

In terms of traffic forecasting, the volatile state of flight schedules and government restrictions create the need for handling a multitude of forecasting scenarios and understanding their associated effect on operations. While difficult, forecasting has never been more important to ensure an effective operation. The ability to forecast the traffic outlook and size the operation accordingly will be the determining factor in a successful and financially sustainable ramp-up and evolving operation.

Aviation Operations during COVID-19 – Business Restart and Recovery
Free Download

This blog post is the first in a series of three where we examine how the restart of aviation will impact the airport operations planning and the importance of agile traffic forecasting. We will provide insight into three cornerstones of the forecasting and planning process – the ones which we believe will lead any airport towards recovery:

  1. How to utilize and upgrade your forecasting capabilities to overcome volatile periods
  2. How to automate processes to limit the amount of manual work in your planning process, increasing efficiency and agility
  3. How to utilize traffic forecasting to optimally size the operation

In each blog post we explore a scenario within forecasting and operational planning. These are scenarios we believe are fueled by the pandemic. For each of these, we will investigate the challenges they pose and their potential solutions. 

Scenario 1: The Double Forecast: In the current environment, airports need to be able to work with a multitude of forecasts. The first step is forecasting if a flight is operating. Secondly, airports need to forecast how many passengers they expect on that flight. We call this challenge the Double Forecast.

Scenario 2: Government restrictions will impact the airport operation: Changing restrictions require the ability to quickly understand the impact on operations and to adjust your planning accordingly.

Scenario 3: Sizing the operation has never been more challenging with the volatility in seasonal schedules: It will be a balancing act between preparing a seasonal forecast and sizing the operation accordingly, while still maintaining agility to adjust continuously.

Scenario 1 of 3: The Double Forecast

We will start by examining why the concept of a Double Forecast is required to overcome the impact of volatility in flight scheduling on airport operations.

Since airlines may choose to operate flights to avoid losing their allocated slots, airports may experience a relatively high number of flights compared to the passenger count. This causes a more diverse downstream impact on the operational areas, where stands and gates will operate at a much higher capacity relative to check-in, security, and immigration.

Although it poses a challenge, we see the Double Forecast as a necessity for all airports. It is challenging since forecasting whether flights are operated or not is difficult and only limited data is available to support assumptions. It is a necessity though, because airports need to understand both flight and passenger aspects even more than before to plan for the downstream impact on operations.

Potential solutions

We recommend an automated forecasting process, with integrations in place to ensure real-time updates of the flight schedule in your forecasting solution. The volatility in traffic schedules requires the ability to quickly update the forecast based on the newest available schedules and passenger numbers. Since the frequency of forecast updates go up, and updates must happen quickly to ensure that the newest passenger numbers are translated into operational impact, integrations and automation is the best way to address this challenge. 

The methodology we recommend is to create a base scenario that includes the flights that have the highest probability to operate. Adding probability to flights is a qualitative exercise where government restrictions, the vaccine outlook etc. are considered, combined with the historical data of how airlines have operated during the previous weeks compared to the original plan.

Working from the base scenario, flights with a lower probability can be added to create additional scenarios, showcasing the potential impact on operation. An alternative method is to utilize a weighted forecast where the probability for each flight and the expected load factor is considered. To manage both approaches, your forecasting methodology needs to be able to handle and forecast individual flights, and to provide the flexibility and granularity needed.


  • Automate your forecasting process
  • Integrate with your available flight schedules
  • Work with probability-based scenarios

Read the next article Restarting aviation: How vaccine passports will change the passenger behaviour.

Rasmus Kaster holds an MA in International Studies from Durham University. Before joining Copenhagen Optimization, he has worked as a military linguist (Persian) for the Danish Armed Forces and with strategy and controlling for the police. He is a firm believer that data, and the optimal visualization thereof, is the foundation for good governance and decision making on all levels of an organization.

Emilie Knutz holds an M.Sc. in Mathematical Modelling and Computation from the Technical University of Denmark, with a specialization in Operations Research and Data Analysis. She enjoys combining and modifying data analytical tools to identify areas of improvement.

The article was provided by a third party and, as such, the views expressed therein and/or presented are their own and may not represent or reflect the views of ACI, its management, Board, or members. Readers should not act on the basis of any information contained in the blog without referring to applicable laws and regulations and/or without appropriate professional advice.

Guest Author
156 articles
Share This