Application of Multiple Regression and Artificial Neural Networks as Tools for Estimating Duration and Life Cycle Cost of Projects

Abstract

Project managers face complex challenges when planning project stages because contract durations and project costs are difficult to predict accurately. The purpose of this study is to investigate statistical tools and concepts that can be integrated in the second phase of the project life cycle: the planning stage. Furthermore, this study aims to compare the accuracy of multiple regression and artificial neural network models, as well as the application of simulation in construction models used in predicting project duration and cost. This paper will also discuss the industry's current estimation methods, the use of statistical approaches, simulation, and the relationship between the application statistical tools and project success. Thus, this review identifies the trending statistical tools used by scholars to develop regression and neural models to solve the complexity of cost and duration estimation. The findings indicate that although the industry needs more accurate predictions and estimating tools, and regardless of the investigations and advancements made with integrating statistical tools, implementing these statistical approaches is faced with barriers.