The ever-increasing demand for transportation of people and goods as well as the massive accumulation of population in urban centers have increased the need for appropriate infrastructure and system development in order to efficiently manage the constantly increasing and diverse traffic flows. Moreover, given the rapid growth and the evolution of Information and Communication Technologies (ICT), the development of intelligent traffic management systems that go beyond traditional approaches is now more feasible than ever. Nowadays, highways often have sensors installed across their range that collect data such as speed, density, direction and so on. In addition, the rapid evolution of vehicles with installed computer systems and sensors on board, provides a very large amount of data, ranging from very simple features such as speed, acceleration, etc. to very complex data like the driver’s situation and driving behavior. However, these data alone and without any further processing, cannot solve the congestion problem. Therefore, the development of complex computational methods and algorithms underpins the chance to process these data in a fast and reliable way. The purpose of this paper is to present a traffic control ramp metering (RM) method based on machine learning and to study its impact on a selected highway segment.