Evaluation of Principal Components When Selecting Subcontractors to Preserve Quality and Reduce Project Time Using Neural Networks

Abstract The increasing momentum of the development process has led to the selection of a suitable contractor as an important factor in the success of projects. Due to the considerable quantity of construction budget in countries that are developing, particular attention should be paid to subcontractor selection to avoid wasting capital, time and reducing quality of work. Subcontractor selection is one of the Basic decisions of managers and decision makers. There is not currently the method of efficient and modern management approach to contractor selection and no attention is paid to scientific methods and techniques. Identifying and evaluating a set of contractor selection criteria will eliminate inefficient contractors from the bidding process. Criteria need to be collected and processed to increase the performance of selected contractors and preparing the needs of the community and the organization's standards. Obviously, there are many qualitative and quantitative indicators to determine the qualifications of contractors, including the ability to do the job, the economic value, the quality, etc. that are important to employers. The purpose of this study is to minimize the losses caused by inappropriate selection of subcontractor and to present a suitable method for selecting subcontractor. In the neural network approach, models for efficient and effective prediction of subcontractor are selected based on the criteria. The most important advantage of radial neural networks over back propagation (BP) conventional networks is the accuracy and speed of approximation in the output of network.