Study on factors affecting the performance of construction projects and developing a cost prediction model using ANN

Construction Industry is one of the very fast growing industries but it also faces many problems which impinge on the performance of their projects. The aim of this study is to identify the factors affecting the local construction projects and analyse them. A questionnaire is prepared from literature review. The questionnaire contains two parts; part A dealing with the general information of the company and the respondent and Part B is subdivided again into different factors like cost, time, health and safety, client satisfaction, community satisfaction factors, productivity factors and environmental factors. The questionnaire was distributed in Chennai, Kerala and Bangalore industries. Each respondent was asked to rank the factors in a range of one to five. The analysis of the response was done using the SPSS software. The top 5 factors affecting the performance of projects were identified as increase in material cost, inadequate supply of labour, incorrect planning, wrong method of estimation, and poor financial control on site. From the various factors studied, cost factors were taken to develop a model using Artificial Neural Network (ANN) to predict the cost of the skeletal system of a building project. Artificial Neural network are self-adaptive method and can capture functional relationships among data even if the underlying relationships are unknown. The neural network toolbox in MATLAB was used to develop and train the neural network models. © 2015 CAFET-INNOVA TECHNICAL SOCIETY. All rights reserved.