Analysis of transportation mode choice using a comparison of artificial neural network and multinomial logit models

The transportation system around the globe is witnessing a dramatic change which possibly generating from the massive increase in the population. This contributed to a legitimate dilemma which is traffic congestion taking into consideration the accompanying problems that raised namely air pollution as well as traffic accidents. Public transportation is substantial and their importance reflects in both economic and social quality of each and every citizen life. Despite these facts, the public means of transportation is still to this day not the people’s choice to perform their daily trips, this applies, particularly, to private car users. The candid solution to this problem is to turn people’s attention to public transportation system (bus and vanpool) and simulate them to abandon their private cars. This study works on a comparison between two mode choice models, Multinomial logistic regression (MNL) and Artificial neural network (ANN) for the purpose of prediction of the behavioural transportation of mode choice with the purpose of evaluation of the accuracy levels in the predictability in each model. The results show that artificial neural network readily outperformed the multinomial logistic regression in the predictability of mode choice.

[1]  Patricia L. Mokhtarian,et al.  What Affects Commute Mode Choice: Neighborhood Physical Structure or Preferences Toward Neighborhoods? , 2005 .

[2]  Ramli Nazir,et al.  An Artificial Neural Network Approach for Prediction of Bearing Capacity of Spread Foundations in Sand , 2015 .

[3]  Stina Adielsson.  Statistical and neural network ananlysis of pesticide losses to surface water in small agricultural catchments in Sweden , 2005 .

[4]  Mohammad Ghodsi,et al.  Comparison of artificial neural network and logistic regression models for prediction of mortality in head trauma based on initial clinical data , 2005, BMC Medical Informatics Decis. Mak..

[5]  O. Bilukha,et al.  Road traffic fatalities in selected governorates of Iraq from 2010 to 2013: prospective surveillance , 2016, Conflict and Health.

[6]  B. Schölkopf,et al.  Sparse Multinomial Logistic Regression via Bayesian L1 Regularisation , 2007 .

[7]  Metin Mutlu Aydın,et al.  M ODELING OFDRIVER LANE CHOICE BEHAVIOR W ITH ARTIFICIAL NEURAL NETWORKS (ANN) AND LINEAR REGRESSION (LR)ANALYSIS ON DEFORMED ROADS , 2014 .

[8]  Mark Beale,et al.  Neural Network Toolbox™ User's Guide , 2015 .

[9]  Abhay Saxena,et al.  ASSESSMENT OF HUMAN CAPACITY BASED ON CONJUGATE GRADIENT TECHNIQUE USING ARTIFICIAL NEURAL NETWORK , 2013 .

[10]  Praveen Kumar Edara.  Mode Choice Modeling Using Artificial Neural Networks , 2003 .

[11]  Michael Green,et al.  Comparison between neural networks and multiple logistic regression to predict acute coronary syndrome in the emergency room , 2006, Artif. Intell. Medicine.

[12]  K. RaghavendraB..  Evaluation of Logistic Regression and Neural Network Model With Sensitivity Analysis on Medical Datasets , 2011 .

[13]  Mohammed Qasim Abdul Ghafoor Al Ani,et al.  Urban Optimization of Transit Oriented Development in Baghdad City , 2016 .