Optimizing Airport Disembarkation Processes: A Novel Approach Using Approximate Bayesian Computation
- Joao Paulo Silva Lima
- Michelle Carvalho Galvão da Silva Pinto Bandeira
- Viviane Adriano Falcão
a Federal University of Goiás (UFG), Faculty of Science and Technology (FCT), Laboratory of · Intelligence and Innovation in Transportation (LI2T) · b Federal University of Pernambuco (UFPE), Department of Civil and Environmental Engineering
- Event
- XXI SITRAER – Fortaleza, CE, Brasil, 2024
- Volume
- Anais do XXI Simpósio de Transporte Aéreo – SITRAER 2024 · ISBN 978-65-272-1008-5 · DOI 10.29327/xxi-sitraer-2024
- Pages
- 250–259
- Publisher
- Even3 Publicações
- Editors
- Francisco Heber Lacerda de Oliveira, Fabiano Rocha, Giovanna Miceli Ronzani Borille, Evandro José da Silva, Mauro Caetano
Abstract
This study develops a macrosimulation model for the disembarkation processes at an international airport, utilizing Approximate Bayesian Computation (ABC) to estimate queue model parameters with intractable likelihood functions. The proposed approach employs ABC with weighted locallinear regression, enabling the inference of marginal posterior distributions for unknown parameters such as the mean pedestrian walking speed and the service rate of immigration queuing systems for both local and foreign passengers. The results demonstrate that ABC effectively transformed prior distributions, even with limited data, into reliable posterior distributions with a 95% confidence interval. This methodological advancement is particularly relevant for the operational management of airport terminals, providing valuable support for future decision-making regarding the capacity and efficiency of queuing systems. The originality of this work lies in the application of ABC within an airport queuing context, where traditional methods fail due to system complexity. By focusing on flight characteristics and immigration passing rates, the study addresses a significant gap in the literature, which has thus far lacked robust approaches for modeling complex queuing systems with intractable likelihood functions. Furthermore, this work paves the way for future applications of ABC in other areas of the airport environment, such as boarding and transfer processes, and suggests integrating other methodologies to further enhance the understanding of these systems.
Keywords: Queue problem; Pedestrian Traffic; Approximate Bayesian Computation; Decision Support