Evaluating and predicting Key Performance Indicators (KIM) facilitates monitoring and controlling project progress. Limited work has been done in forecasting construction project performance using Key Performance Indicators. Developing key performance indicators that model the project performance over its life cycle provide useful management tools. This paper proposes a model to forecast the construction project performance and introduces a quantitative method to measure KPIs of projects dynamically. Key performance indicators assess different aspects of projects and are used as a thermometer to demonstrate the health status of projects. The first step in measuring project performance is defining the project objectives and clarifying from whose point of view the performance is measured. Then, indicators are measured by considering the profits and damages of the specific stakeholder. For quantitative KPI, mathematical calculations are applied and for qualitative indicators, a questionnaire is designed and used. The relative weight factors of each indicator are determined using Analytic Hierarchy Process (AHP). Subsequently, mathematical calculations are performed to get the project overall performance. This paper applies neuro-fuzzy technique to develop the model for forecasting construction project performance. The proposed method uses a set of key performance indicators to predict the project status at different time horizons. Then if the project has a major deviation, corrective actions arc used to improve project performance. This model can be used in building construction projects to help decision makers evaluate and improve the performance of their projects.