Machine Learning Based Model for Traffic Prediction in Smart Cities

Abstract

A smart city can be defined as a city that uses distributed sensors to monitor and control the urban environment by collecting real-time information. Using such smart devices improves the quality of living and facilitates the process of decision making. Smart cities provide solutions to environmental issues such as amount of wastage, energy consumption, traffic congestion, and pollution. The traffic issue is a one important concern in any city due to the increasing number of populations using vehicles which lead to traffic congestions, accidents, and delays. Traffic issues can also cause a high level of pollution and fuel consumption. To solve such issues, roads should be managed by monitoring, analyzing and predicting the traffic flow. In this research, we propose a machine learning based model for traffic flow prediction in smart cities, particularly in the context of “NEOM” megacity that is born from Saudi Arabia's vision of 2030. The proposed model combines previous methods of traffic prediction to produce robust traffic prediction model that can be used in NEOM megacity to achieve better traffic management.