Multi orthogonal review of modern demand forecasting lines and computational limitations in Green Urban mobility

Abstract

Urban mobility attempts to combine payment systems asa service with mobility, which has been divided into several transportation segments, and offer door-to-door services to consumers. Demand forecasting in the transportation sector is usually done in pairs, based on origins and destinations. To be more precise, forecasts are made for the volume of container traffic, vehicle traffic, and passenger departure and arrival. The purpose of this work is to examine the literature on demand prediction forecasting in several transportation domains, including vehicle sharing, leased cars, bicycles, and public transportation. The novel assessment preferred research papers to applied machine learning, deep learning, neural networks and Quantum learning methods. The study justifies the difference between Quantitative and Qualitative demand prediction. This review examined in different levels such as forecasting methods, hybrid models and quantum machine learning methods. Each existing research works classified into algorithms, prediction and observed results in numerical. Finally, the survey effort to find the strengths and limitation of the prevailing past research approaches.