A COMPREHENSIVE REVIEW FOR ARTIFICAL NEURAL NETWORK APPLICATION TO PUBLIC TRANSPORTATION

Abstract

This paper presents a comprehensive review of research studies related to the application of artificial neural networks (ANNs) to public transportation (PT) since 2000. PT applications with ANNs have a great prominence because it provides an opportunity of prediction, comparison and evaluation in PT. A short introduction for applied studies in public transportation based on NN is included to guide the unfamiliar readers and a detailed review table has been presented in the paper. More than a thousand studies have been viewed, however, 72 studies of PT are related to ANN. It is observed that multi-layer feed forward network with gradient descent training has been commonly used by now. In contrast, the other less known methods are prone to increase. This paper guides future research directions and presents the methods to be exerted in PT for input determination.