We've all been in a situation where we were wondering while sitting in a congestion: why should I wait for the red light to pass through the intersection without any trouble? And indeed, when designing traffic light programs, engineers sometimes have to give priority to the main directions of progress. But what if there is a compromise solution for a safe and fast passage to leave the intersection? There are millions of ways to set up an intersection according to which track you want to pass, how many seconds to give for the green lights, etc. Machine learning will help you. We'll tell it how the junction looks and where the congestion is and let it try to combine possible solutions until the best option is set. The machine then comes up with a program that already meets the safety standards and provides the most practical method for every person in traffic.

Once a junction or intersection changes, you need to redesign the traffic light program and design. As the geometry of the intersection changes, new unforeseen problems may arise. Machine learning can quickly, even within hours, come up with an optimal solution. This means that if required, the machine will produce standard programs, and if it is a road section, it will be able to implement a coordinated system of the entire road section. The tuning is made so that vehicles can receive green waves in several directions, and traffic can be continuously and safely controlled. For people this is a job for several days, but the machines are only a few hours now. In addition, documentation is done in standard format automatically.

Each city is unique and the expectations of the environment can be different. The image of the city is greatly influenced by how good the organization of transport is. Traffic is changing with the development of the city. Since the number of vehicles is constantly increasing, it may be important that vehicles pass faster on some road sections. For example, after a major event, it is important that those present are able to leave the scene quickly without disturbing the main drivers of the city's traffic. Machine learning can be used in traffic management in the following ways:

• emptying the inner areas of the city

• serving morning incoming traffic

• quick evacuation programs after events

• quickly compensate for congestion due to road rebuilding or road closures by redirecting traffic

• Any other traffic acceleration or capacity transfer