The project addresses the evolution of cellular networks, currently in the fifth generation (5G), and anticipates the next generation (6G) planned for 2030. 5G networks have three main use cases: enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communications (mMTC). These use cases introduce new management challenges due to the different requirements for latency, data rate, and packet loss. To meet these requirements, technologies such as millimeter-wave transmission and network slicing have been incorporated. However, these advances increase the complexity of network management. To address these challenges, machine learning and artificial intelligence techniques have been explored, as in previous projects by UFPA and Ericsson. Another important concept is the digital twin, a digital replica of the real environment to test solutions before implementing them. The project aims to evaluate and develop solutions based on digital twins, machine learning, and artificial intelligence to improve cellular network management. This includes enhancing the existing experimental cellular network, implementing a digital twin of relevant parts of the network, and using this digital twin to optimize processes. The proposed activities aim to solve important problems for the development of 5G and 6G networks, in addition to advancing machine learning techniques and data generation for model training.