The construction industry is subject to a high level of risks and uncertainties than any other industry. In reality, most participants experience risks in cost and time overruns and often fail to meet quality standards and operational requirements. In order to overcome these risks and make decisions with high accuracy, traditional and smart techniques have been applied to predict the cost and delay of construction projects with a high degree of accuracy and minimal errors.
This research aimed to investigate the accuracy of five Artificial Intelligent techniques (Artificial Neural Network, Support Vector Machine, Extreme Learning Machines, Artificial Neural Network-Particle Swarm Optimization, and the Adaptive Neuro-Fuzzy Inference System) to demonstrate the impact of risk factors on prediction the cost and delay of construction projects. These techniques were represented through a virtual graphical user interface allowing the user to ease and clarify use.
This study collected data from 47 construction projects from the AL-ZAWRAA state company in Baghdad city. Thus, the factors of risk were specified as well as analyzed employed Probability and Impact Analysis which were adopted as the inputs of the models. In contrast, the outputs of models were represented contractor’s profit ratio to the costs of project and the delay in the construction project.