Análise de desempenho de algoritmos de aprendizagem de máquinas para análise desagregada de viagens intermunicipais

Authors

  • Andreza Dornelas de Souza Roma Universidade de São Paulo
  • Cira Souza Pitombo Universidade de São Paulo
  • Henrique Stramandinoli Guimarães Universidade de São Paulo
  • Luis Henrique Magalhães Costa Universidade Estadual Vale do Acarau

DOI:

https://doi.org/10.14295/transportes.v26i3.1614

Keywords:

Trip Distribution, Genetic Algorithms, Decision Tree, Gravitational Model.

Abstract

This paper proposes a disaggregated analysis of intercity destination choices, through the application of Machine Learning (ML) algorithms (Classification And Regression Tree - CART and Genetic Algorithms - GA). An Origin-Destination Survey was carried out by the Center of Transportation and Environmental Studies (UFBA) in 2012/2013 in eleven municipalities in the state of Bahia, Brazil. It was carried out a calibration of a Multinomial Logit Model with GA algorithm, bringing the advantage of association of the destination choices to values of estimated coefficients of the random utility functions, without the problems related to the calibration of the traditional logit models, such as Irrelevant Alternatives (IIA) assumption. The performance of each ML algorithm was compared to a traditional approach (Gravitational Model). The results showed that the ML algorithms presented better predictions for destination choices, and GA presented advantages in obtaining the estimated parameters related to the covariates. The main conclusion is that such algorithms can be applied in trip distribution step, incorporating the effect of the disaggregated variables, without rigorous assumptions of the traditional disaggregated models.

Downloads

Download data is not yet available.

References

Ben-Akiva, M.E.; Lerman, S.R. (1985) Discrete Choice Analysis: Theory and Application to Travel Demand. The MIT Press, Cam-bridge, MA.

Breiman, L.; Friedman, J.H; Olshen, R.A.; Stone, C.J. (1984) Classification and Regression Trees. Wadsworth International Group, Belmont, CA.

Carvalho, A. C. P. L. F.; Galvão, C. O.; Lacerda, E. G. M.; Diniz, L. S.; Valença, M. J. S.; Ludermir, T. B.; Vieira, V. P. P. B. (1999). Siste-mas inteligentes: Aplicações a recursos hídricos e ambientais. Porto Alegre: Editora Universidade/ UFRGS/ ABRH. ISBN 8570255276.

De Grange, L.; Fernández, E.; de Cea, J. (2010) A consolidated model of trip distribution. Transportation Research Part E: Lo-gistics and Transportation Review, v. 46, n. 1, p. 61–75. DOI: 10.1016/j.tre.2009.06.001

De Souza, A. D. (2017); Comparação de algoritmos de Aprendizagem de Máquinas para análise desagregada de viagens intermu-nicipais. 84 f. Dissertação de Mestrado. Departamento de Engenharia de Transporte. Escola de Engenharia de São Carlos.

Fotheringham, A.S. (1983) Some theoretical aspects of destination choice and their relevance to production-constrained gravity models. Environment and Planning A, v. 15, n. 8, p. 1121–1132. DOI: 10.1068/a151121

Ichikawa, S.M., Pitombo, C.S., Kawamoto, E. (2002) Aplicação de Minerador de dados na obtenção de relações entre padrões de viagens encadeadas e características socioeconômicas. Anais do XVI do Congresso de Pesquisa e Ensino em Transportes, Anpet, Natal (RN), v. 2, p. 175-186.

Kass, G.V. (1980) An exploratory technique for investigating large quantities of categorical data. Applied Statistics, v. 29, p. 119–127. DOI: 10.2307/2986296

Kononenko, I; Kukar, M. (2007) Machine Learning and Data Mining: Introduction to Principles and Algorithms. Horwood Pub-lishing. Chichester, UK.

Koppelman, F. S.; Wen, C.H. (2000) The paired combinatorial logit model: properties, estimation and application. Transporta-tion Research Part B: Methodological, v. 34, n. 2, p. 75-89. DOI: 10.1016/S0191-2615(99)00012-0

LaMondia, J.; Snell, T.; Bhat, C.R. (2009) Traveler Behavior and Values Analysis in the Context of Vacation Destination and Travel Mode Choices: A European Union Case Study. Transportation Research Record: Journal of the Transportation Re-search Board, n. 2156, p. 140-149. DOI: 10.3141/2156-16

Marsland, S. (2009) Machine Learning: An Algorithmic Perspective. CRC Press. Cambridge, UK.

Mozolin, M.; Thill, J.C.; Linn, U.E. (2015) Trip distribution forecasting with multilayer perceptron neural networks: A critical evaluation. Transportation Research Part B: Methodological, v. 34, p. 53-73. DOI: 10.1016/S0191-2615(99)00014-4

Omrani, H. (2015) Predicting travel mode of individuals by machine learning. 18th Euro Working Group on Transportation, EWGT 2015, p. 840-849.

Pitombo, C.S.; Kawamoto, E.; Sousa, A.J. (2011) An exploratory analysis of relationships between socioeconomic, land use, activity participation variables and travel patterns. Transport Policy, v. 18, p. 347-357. DOI: 10.1016/j.tranpol.2010.10.010

Pitombo, C.S.; Kawamoto, E.; Sousa, A.J. (2013) Linking activity participation, socioeconomic characteristics, land use and travel patterns: a comparison of industry and commerce sector workers. Journal of Transport Literature, v. 7, p. 59-86. DOI: 10.1590/s2238-10312013000300004

Pitombo, C. S.; De Souza, A.D.; Lindner, A. (2017) Comparing decision tree algorithms to estimate intercity trip distribution. Transportation Research Part C , v. 77, p. 16-32. DOI: 10.1016/j.trc.2017.01.009

Pulugurta S, Arun A, Errampalli M (2013) Use of Artificial Intelligence for Mode Choice Analysis and Comparison with Tradi-tional Multinomial Logit Model, Procedia - Social and Behavioral Sciences, v. 104, p. 583-592. DOI: 10.1016/ j. sbspro.2013.11.152

Quinlan, R. (1983) Learning efficient classification procedures and their application to chess end-games. Machine Learning: An Artificial Intelligence Approach, Tioga, Palo Alto, p. 463-482.

Rasouli, M.; Nikraz, H. (2013) Trip Distribution Modelling Using Neural Network. Transport Research Forum, Brisbane, Aus-tralia.

Wilson, A.A. (1967) Statistical Theory of Spatial Distribution Models. Transportation Research, v. 1, p. 253-269. DOI: 10.1016/0041-1647(67)90035-4

Xie, C.; Lu, J.; Parkany, E. (2003) Work travel mode choice modeling with data mining: decision trees and neural networks. Transportation Research Record: Journal of the Transportation Research Board, n. 1854, p. 50-61. DOI: 10.3141/1854-06

Yang, C.; Tsai, M.; Chang, C, 2014. Investigating the joint choice behavior of intercity transport mode and high-speed rail cabin with a strategy map. Journal of Advanced Transportation. DOI: 10.1002/atr.1264

Published

2018-11-04

How to Cite

Roma, A. D. de S., Souza Pitombo, C., Guimarães, H. S., & Costa, L. H. M. (2018). Análise de desempenho de algoritmos de aprendizagem de máquinas para análise desagregada de viagens intermunicipais. TRANSPORTES, 26(3), 159–175. https://doi.org/10.14295/transportes.v26i3.1614

Issue

Section

Artigos Vencedores do Prêmio ANPET Produção Científica