Project Portfolio Management is very important for the growth of companies, because it favors to plan several possibilities in each scenario. The purpose of the Project Portfolio Management is to manage all resources in order to plan and execute successful projects and achieve the strategic objectives of the organizations. In the Project Portfolio Management, a lot of data is generated daily, which is important for the planning of new projects in companies; consequently, this need arises to create models that help to process and interpret this data. In this context, Machine Learning as an expression of Artificial Intelligence, is presented as an alternative and technological enabler that allows a system, by itself and in an automated way, to learn to discover patterns, trends and relationships in data, it is presented as an engine of digital transformation of business, which is being adopted by many organizations and its demand is growing. Therefore, this paper aims to compile and review the proposals made for the implementation of Machine Learning and critical success factors to improve Project Management, based on a literature review and an analysis of the current state of the art of Machine Learning. 122 articles were found and 21 articles were selected that are related to the research questions. As a final result, 7 ML methods and 18 critical success factors for PPM have been identified.