This thesis is about simulating and calculating how traffic patterns and road capacity can be improved using adaptive cruise control to eliminate human reaction times. The thesis will focus on a concept known as phantom queues, and find out if those could be eliminated with the help of adaptive cruise control (ACC). The experiment is conducted with self-driving cars driving around a closed circuit in a queue. The cars in this project were designed in Fusion 360, 3D-printed and assembled using an Arduino as controller. The first car is made to brake and the effect on the queue is observed. This is repeated with different reaction times and distance settings for the ACC to find the optimal distance between cars in order to avoid phantom queues. The optimal distance is compared for different simulated reaction times. The result shows that the capacity of a road could be increased by 40 – 70%, without impeding the flow of the traffic, if adaptive cruise control were to be mandatory in every car. This is because of the decreased distance needed between each car. Finally, a theoretical calculation using data from the specification of a Tesla car shows that it could be possible to achieve even higher rates in a real world setting.