Benefit and cost analysis: Three-dimensional computer models with integrated databases in the management of construction

The purpose of this research is to provide simple, cost effective but structured and robust methodologies for benefit-cost analysis. Two paradigms are investigated. One is classical statistical technique-multiple regression, the other is modern artificial intelligence technique-neural networks. This study shows that artificial neural networks deliver better performance in terms of accuracy. However, regression analysis provides the baseline for the newer methods. Each has unique advantages and disadvantages, and hence each is appropriate for distinct types of problems. The combination of statistics and artificial neural network yield the best results. The objective of this research is to quantify the benefit-cost impacts that occur on a construction project if three-dimensional computer models with integrated databases are used in the management of construction. The results show that 3D computer models when used properly in the management of construction have significant influence on construction cost and rework metrics. The use of 3D models has little significant effect on construction duration metrics. The level of model use is critical since poor use of 3D computer models is worse than no use at all. This research shows that three-dimensional computer models with integrated databases can significantly improve the construction metrics.