Improve the accuracy of project construction forecasts by integrating managed value management with quantitative predictive methods in time series. Case study: Construction projects at AzarAB Company Arak

In this research, two projects were considered during the construction phase of Azar AB Company to predict construction costs and actual costs of these projects were calculated after data collection, then six regression models were presented in four time periods for each project and then The average model error for each project was calculated in four periods to determine the final and acceptable model for each project. The results showed that models 2 and 3 percent have lower and acceptable errors than other models. The existence of effective modules in models as well as in terms of the nature of projects, we conclude that Model No. 3 predicts the future costs of projects with a reasonable degree of certainty. In sum, we will select Model No. 3 as the top model to predict the future costs of the projects in question. However, we recall that for each project, depending on its nature, another particular equation may provide more accurate predictions. But in general it can be said that regression models are used to predict the costs of different periods The project is applicable to its nature.