Estimate Final Cost of Roads Using Support Vector Machine

Abstract

Introduction: The cost overrun in road construction projects in Iraq is one of the major problems that face the construction of new roads. Method: To enable the concerned government agencies to predict the final cost of roads, the objective this paper suggested is to develop an early cost estimating model for road projects using a support vector machine based on (43) sets of bills of quantity collected in Baghdad city in Iraq. As cost estimates are required at the early stages of a project, consideration was given to the fact that the input data for the support vector machine model could be easily extracted from sketches or the project's scope definition. Result: The data were collected from contracts awarded by the Mayoralty of Baghdad for completed projects between 2010-2013. Mathematical equations were constructed using the Support Vector Machine Algorithm (SMO) technique. Conclusion An average of accuracy (AA) (99.65%) and coefficient of determination (R2) (97.63%) for the model were achieved by the created prediction equations.