An Improved Method for Project Duration Forecasting
Xiaoxiao Chen, Lu Liu, You Li · 2010
There are many factors affect the accuracy of project duration forecasting, the lack of relative information and the complexity of the project are two major aspects. To overcome these constraints and establish a feasible forecasting model, this paper presents an improved method to forecast the project duration, which combines the earned schedule and artificial neural network. We adopt the artificial neural network because of its ability to model complex nonlinear relationship without a priori assumption of the nature of the relationship. The performance of the developed models was evaluated. The results show that the artificial neural network can improve the accuracy of the project duration forecasting significantly in terms of the error evaluation measurements.