Intelligent optimisation system for airport operation: Hajj Terminal in Saudi Arabia

Abstract

Airport operation level of service (LOS) and performance management are among the major concerns by any airport authority. Two aspects considered in that kind of measurement: passengers prospective and operators prospective. This thesis tries to combine both in its produced optimisation system. This study was carried out in the Hajj terminal of the King Abdul-Aziz international airport and classified the processing time among the most important measures affecting the users’ observation of the level of service. Produced survey has helped to generate performance measure upon passengers prospective. On the other hand a simulation model of the process flow is utilised to formulate driven data model of the terminal process flow operations. The model built on Arena software and correlation study is made from the multiple “what if” scenarios of the model. Then a linear regression is used to generate a model for each variable. Levenberg–Marquardt (LM) algorithm is used after to carry out better regression model then Neuro-Fuzzy (NF) model found to be more efficient as it is picked and used to generate a best observed prediction. The system is optimised through the generated Neuro-Fuzzy (NF) logic model using both Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). A validation in addition to the testing made in the optimisation system. Analysis shows a great deal of improvement in predictions using fuzzy logic instead of linear regression for all dependent variables. PSO and GA optimisations are carried out and compared to the actual results gathered from the Arena simulation report.