The completion of projects within the limits of the planned cost is an important component of projects management. Therefore, estimating the cost at completion of projects is a difficult task, Nevertheless the developed of technological tools will enable the planner to better understand the cost estimation processes in the various stages of construction projects. The objective of this study is to improve the predictive performance of earned value management by developing a model to estimate the cost at completion of construction projects using artificial neural network technology. In this research, two techniques were used to estimate the cost at completion of construction projects, the first technique is multiple linear regression, and the second technique is artificial neural networks. The Bus Rapid Transit project in Amman was chosen to apply the research problem as a case study. The data of the project consisted of eight packages, which included (207) monthly follow-up reports were collected. The three basic values of the earned value management (PV, AC, and EV) were used as inputs to the models to estimate the cost at completion (EAC). The results showed that the ANN technique is better than the MLR technique for estimation cost at the completion of the construction projects based on the mean absolute percentage error and average accuracy rate, which was equal to 5.93%, and 94.07% respectively for the ANN model, and in MLR model equal to 31.61%, and 68.39% respectively. These results indicate that the ANN technique should be used in estimation cost at completion of the construction projects, as it saves a lot of effort and time for decision-makers and project managers to take the appropriate decision.