Affiliation:
1. Department of Civil Engineering, University College of Engineering, Thirukkuvalai, Nagapattinam, Tamilnadu, India
Abstract
As the population increases day by day, there has been a tremendous increment of ownership over two-wheeler and four-wheeler systems. Thereby usage of public transports system decreases than private vehicles system. In this scenario, this study focused on the significant factors in the mode choice behaviour of commuters from the extensive household survey conducted by Thiruvarur City to obtain detailed information on the modal split pattern analysis. The questionnaire of 4857 respondents was used in the Multinomial Logit Model using the Statistical Packages for Social Science (SPSS) tool to predict the mode choice behaviour of commuters. A total of seven explanatory variables were extracted from the responses; the Multinomial logit model was calibrated to fit the data, and the chi-square test was used to measure the goodness of fit. Apart from the MLM model, commuter behaviour prediction analysis is also made using soft computing technologies, which involves multi-linear regression and neuro-fuzzy model, keeping uncertainties involved in commuter behaviour to avoid collecting household data. It was found that two-wheeler commuters are more in numbers for their travel choice as cost minimizing than four-wheelers, public transport system commuters. Consequently, the developed model predicts the behaviour of commuters in a precise manner that closely matches the actual conduct.
Subject
Artificial Intelligence,General Engineering,Statistics and Probability
Reference42 articles.
1. Representing Automated Vehicles in a Macroscopic Transportation Mode;Török;Periodica Polytechnica Transportation Engineering,2020
2. Transport modeling with a purpose: How urban systems design can bridge the gaps between modeling, planning, and design;Binder;Proceedings of the Urban Transport, 2019, 25th International Conference on Urban Transport and the Environment, Aveiro, Portugal
3. Machine Learning Travel Mode Choices: Comparing the Performance of an Extreme Gradient Boosting Model with a Multinomial Logit Model. Transportation Research. Record;Wang;Journal Transportation Research Board,2018
4. Values, attitudes, and travel behaviour: A hierarchical latent variable mixed logit model of travel mode choice;Paulssen;Transportation,2014
5. Inclusion of latent variables in mixed logit models: modelling and forecasting;Yáñez;Transportation Research Part A: Policy and Practice,2010