POSITIONING PREDICTION FOR PASSENGERS QUANTITY OF INDONESIAN DOMESTIC FLIGHTS USE ARTIFICIAL NEURAL NETWORK AND MINIMUM SPANNING TREE

• Main Author: MEYLANI (NIM: 10113058), RENY
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/23896
• Institution: Institut Teknologi Bandung

The similarity of passengers quantity on an airline route is crucial to the policies adopted by airline companies, airport companies and government. The movement of domestic flight passengers can be assumed by the movement of passengers from every city in Indonesia to Jakarta, because this town has connectivity with more cities in Indonesia. Therefore, data of the number of passengers from all cities in Indonesia to Jakarta in 2013 to 2016 is used to study the dynamics of similarity of passenger density among cities in Indonesia. The prediction of passengers quantity is done using the artificial neural network method and in order to visualize the resemblance of the dynamics of the passenger quantity, the minimum spanning tree is used. Predicted results show in 2017 the passenger quantity of flights to Jakarta will decrease and the value of dynamics correlation movement of passengers for each city in Indonesia decreases every year. Through position maps, it can be concluded that each city has a difference in the relationship relevance of passenger density dynamics each year, but cities in the same island tend to have a high proximity.