This article focuses on improving the business tracking system by focusing on deadline, costs and resources. The main objective is to use Artificial Intelligence to develop a robust prediction model to facilitate the effective monitoring of projects within companies. The study begins with the collection of information on traditional project monitoring methods, based on the use of predefined and static indicators to assess progress in the three specific aspects of each project. However, these indicators fail to capture the real complexity and dynamics of projects. To obtain more effective monitoring, a study is carried out using two learning models: the first based on the algorithm Support Vector Machine and the second based on the multi-layer perceptron’s. An in-depth analysis of the problems of the existing monitoring system is carried out, followed by the collection and preparation of historical data sets from previous projects. Relevant characteristics related to time, costs and resources are carefully selected for analysis. To classify projects according to their progress, the Support Vector Machine and the Multi-layer perceptron’s are applied to the data set to obtain a reliable classifier. Finally, a comparative study of the results obtained by the two algorithms is presented. The ultimate goal is to improve project monitoring using Artificial Intelligence-based methods, which should allow more efficient management of time, costs and resources for the whole company.