Research on Project Duration Prediction Based on Multiple Linear Regression

In the engineering management of power grid construction projects, effective prediction of construction process durations is one of the critical factors affecting the quality and cost of the project. Based on historical infrastructure project data, this paper establishes a multiple linear regression model to predict the expected completion time of each process of the current engineering project, using the main influencing factors of project process duration as the independent variable, duration as the dependent variable, and the calculated rated duration as the constant. Using network planning techniques and analyzing the project Gantt chart, the network node diagram of the project is drawn, and the critical path of the project execution is determined. Finally, based on the fuzzy membership function duration calculation method, the total project duration prediction range is obtained.