Evaluating the impact of machine learning with regards to customer experience within airports : a study in the local context

Abstract

The airline industry is one of the most technologically advanced sectors providing air transport to both people and cargo. One of the main aims of an airport is to retain a high level of customers’ experience. This dissertation aims to investigate the possibility of using machine learning to predict the number of passengers on a flight thereby helping Airport Management with the allocation of their workforce and resources.

The data acquired from Malta’s International Airport is sorted and cleaned according to the conducted research in order to make it much more valuable to the learning algorithms. These are used to generate predicted ‘Levels’ for every instance. The model is then trained and tested against unseen data sets to determine the accuracy of the classifier. Subsequently, the classifiers are sorted, from the most accurate to the least accurate.

A prototype is done using WEKA library where the Airport Management can enter the attributes of a particular flight and obtain a prediction of how full the flight would be. This would in turn aid the management to perceive the number of passengers the airport would have on a particular date and ensure that they have the necessary workforce to give the best customer experience.