Application of committees of neural networks for conceptual cost estimation of residential buildings

Abstract

The paper presents concisely some research results on the application of committees of artificial neural networks (that is committees of networks also called committee machines) for conceptual cost estimation of residential buildings. The author focused on application of chosen static structure type of network committees in regression problem binding together the basic information about residential buildings available in the early stage of design and construction cost. The goal of the research was to improve the formerly proposed regression model based on a single network – especially to minimize the number of occurrences of errors with a high value. Due to the results, in the described problem of conceptual cost estimation obtained, committees of networks proved to be better solution for the regression model than a single networks. The conclusion is that a neural approach involving committees of artificial neural networks may be an alternative both for the single neural network based models and the traditional methods of conceptual cost estimation in construction projects.