And the forecast is … more passengers travelling by air
The number of people travelling on aircraft in Europe continues to grow each year. This annual increase will add to the challenges already faced by airports and airlines which must be able to plan their complex operations in advance. However, these well laid plans often have to be altered at the last minute due to changes in local circumstances, which can have a cumulative impact across Europe. Therefore, high performance forecasting tools and systems are required for predicting the movement of passengers, enabling airports and airlines to optimise their resources accordingly. Passenger movements can be influenced by a range of often random factors, making it difficult to provide accurate models. Researchers attempting to forecast passenger movements have turned to algorithms used in data mining to sift through huge numbers of events to identify a possible relationship. The technique is combined with text mining of relevant information from the internet in a number of languages to provide powerful new forecasting tools. The Airforce project developed methodologies to enable the automatic extraction of sets of important parameters from a sea of data and extrapolate results on departure data with only limited information. The consortium also developed the necessary infrastructure for retrieving useful information from established databases in order to supply the forecast server developed by Airforce. A database was also created and constantly updated in order to gain a better understanding of travel trends. These trends included who is travelling and when, and which areas are experiencing an increase in passenger numbers. Different prototypes developed by Airforce were tested for the accuracy of their forecast and the results compared with the systems currently used by airlines and airports. The project's results can be used in a range of related areas and can help to ensure the competitiveness of European companies employing this technology. Reports from end-users have been positive and show a clear interest in the system.
