Prediction of Passenger Count at Halu Oleo Kendari Airport Using Multi-layer Perceptron

Abstract

Transportation, particularly air travel, plays a vital role in supporting economic and social activities of the society. In this context, predicting the number of airline passengers becomes crucial for airport planning and management. This research aims to utilize the Multilayer Perceptron (MLP) method to predict the passenger count at Halu Oleo Kendari Airport. The MLP method is a type of neural network that exhibits high predictive capabilities. In this study, train score and test score are used to evaluate the performance of the MLP model in predicting the training and testing data. Additionally, Mean Absolute Percentage Error (MAPE) is employed to measure the accuracy of the model's predictions. The results indicate that the train score of 60% signifies the model's ability to predict the training data with an accuracy of 60%. Meanwhile, the test score of 96% demonstrates the model's good predictive capability on unseen testing data. The MAPE value of 0.152% indicates a relatively low level of error in the model's predictions.