The Genetic Algorithm-Artificial Neural Networks Integration in the Optimization: An Application for Transportation Systems

Çaparoglu Ö., Ok Y., Özaydin N.

INTELLIGENT AND FUZZY SYSTEMS, INFUS 2024 CONFERENCE, VOL 1, Çanakkale, Turkey, 16 - 18 July 2024, vol.1088, pp.277-284 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 1088
  • Doi Number: 10.1007/978-3-031-70018-7_30
  • City: Çanakkale
  • Country: Turkey
  • Page Numbers: pp.277-284
  • Keywords: Artificial Neural Network, Genetic Algorithm, Transportation, Nature-Inspired Optimization Algorithms, Optimization, Machine Learning, Greenhouse Gas Emission
  • Ataturk University Affiliated: Yes

Abstract

The transportation sector stands as one of the broadest and most significant industries worldwide. Today, one of the key environmental consequences of the transportation sector is the increase in carbon dioxide (CO2) emissions. This research focuses on CO2 emissions, specifically from road transportation. The primary objectives of the study are to investigate how yearly fluctuations on highways impact CO2 emissions. The study utilizes genetic algorithm-based artificial neural networks (GA-ANN) to design a transportation optimization model. Artificial neural networks are employed to estimate the proportion of CO2 emissions from roadways in the total transportation-related CO2 emissions. The article proposes an optimization method using genetic algorithms (GA) to optimize the parameters of the artificial neural networks. The development of the ANN model was carried out out in the MATLAB program. The constructed GA-ANN model was compared to a variety of other machine learning (ML) models, such as linear regression (LR), support vector machine (SVM), random forest (RF), and artificial neural network (ANN).