Public–private partnership, “PPP”, is considered the new era in construction projects in developing countries. According to the World Bank, public–private investment in the infrastructure industry has accounted for US $76.2 billion in 2021, compared to US $51 billion in 2020, which represents a global increase of 49% from 2020. Moreover, according to the MENA forum, more than 32,176 infrastructure projects with a total value of US $3 trillion are being implemented in the Middle East. Governments tend to promote PPP projects in order to make full use of the technical and financial capabilities of the private sector to be able to enhance the social and economic level of the country for the benefit of the public. A major critical success key factor in such projects is proper risk allocation between the private and the public partners. Among such risks is the accurate identification of the concession period of the project. Proper concession period is considered a win–win situation for both the public and private partners, as long concession periods will be in the benefit of the private partner while short concession periods will be in the benefit of the public partner. One of the major factors affecting the identification of the concession period is predicting the rate of inflation across the project life cycle. The goal of this paper is to develop and validate a machine learning model with the capability of forecasting future inflation rates, which is then utilized in the optimization of the concession period. Historical data that includes several construction factors affecting the inflation rate is used to train and validate the prediction model. Vector Autoregression (VAR) is employed to produce this predictive model. Upon validating this forecasting model, the predicted inflation rate will be utilized in the optimization of the concession period, along with other risk factors for PPP projects.