One of the main challenges in optimizing services in various fields, including the legal sector, is the lack of an automatic system that can efficiently assign tasks based on predefined rules. In many cases, documents are still processed manually by qualified staff, which can be time-consuming and prone to errors. To address this issue, there is a need to develop a system that can automate the classification process, thereby simplifying the allocation of tasks to the most competent professionals in a specific field. This can be achieved by using Natural Language Processing (NLP) algorithms. In our Case Study the problem is to classify legal documents and suggest the most suitable judiciary branch for each dossier. By implementing such a system, it is possible to reduce processing times and enhance the overall efficiency of judicial offices. The proposed method can be supported by using a legal distributed system, which can facilitate seamless communication and collaboration among different actors.