The Integrated Methodology of Rough Set Theory and Artificial Neural-Network for Construction Project Cost Prediction

Abstract

Construction cost estimation and prediction, the basis of cost budgeting and cost management, is crucial for construction firms to survive and grow in the industry. The objective of this paper is to presented a novel method integrating rough sets (RS) theory and anarticial neural network (ANN) to forecast construction project cost. Becouse there are many factors affecting the cost of building and some of the factors are related and redundant, rough sets theory is applied to find relevant factors to the cost, which are used as inputs of an articial neural-network to predict the cost of construction project. Therefore, the main characteristic attributes were withdraw, the complexity of neural network system and the computing time was reduced, as well. A case study was carried out on the cost estimate of a sample project using the model. The results show that the integrating rough sets theory and artificial neural network can help understand the key factors in construction cost forecast, and provide a way for projecting more reliable construction costs.