Automated Cost Estimation in Construction Using Ai Approach

Abstract

A number of cost estimation methods have been observed for distinct projects in the state-of-art like as estimation using simplified distribution cost estimation, valuation using cost functions such asbased on activity, index method and many more. The conventional cost estimation techniques follow quantity, and comparative cost estimation, which is based on different factors such as project type, area, and volume of site. When traditional cost estimation methods are used in the projects the specifications are assumed that cost estimates provide a direct relationship between the ultimate cost and the main design attributes of theconstruction site. However, the hypothesis of a direct relationship is objectionable. With the increase in the technology in computer and software applications, it facilitated new techniques to estimate cost with a little human effort. In this research, Artificial Neural network (ANN) is used as an Artificial Intelligence (AI) approach, having the ability to resolve the non-linear problems and provide results with higher accuracy. This research examines the benefits of ANN approach to overcome the problem of cost estimation in the initial step of structure design. The rate and Bill of Quantity (BoQ) of 10 educational projects are used to learn and validate ANN based model. Around 10 attributes are fed at the input layer to train the designed model. The design parameters used to estimate the cost per square meter of educational building in India with an average accuracy of 97 % was obtained.