Support Vector Machine to Measure the Risk of Payment Delay on Construction Projects in Gaza Strip

Abstract

Payment delay to contractors is a common cause of disputes in Gaza Strip construction industry. The aim of this paper is to formulate a model to measure the risk of payment delay in terms of cost overrun. Factors that contribute to the risk of payment delay were identified. 140 questionnaires were distributed to contractors. 113 (80.71%) questionnaires were received. Nine factors were identified to be the most important factors that are contributing to the risk of payment delay. A model to measure the risk of payment delay was formulated. Support Vector Machine (SVM) was used to develop the model. A hypothetical case study and structured interview with (31) contractors was used to build up the model. The nine factors that contribute to the risk of payment delay were used as input data and the total overrun cost due to payment delay was used as an output. The proposed model was able to predict the cost overrun due to payment delay with accuracy performance of 93.47%. The model can help to give early warning of cost overrun due to payment delay. Keywords Payment Delay, Modeling, Support Vector Machine, Gaza Strip.