Modelling cost-flow forecasting for water pipeline projects using neural networks

This paper furthers work that already exists in the use of artificial intelligence techniques to forecast cost flow for construction projects. The paper explains the need for cost-flow forecasting and investigates the methods currently used to perform such a task. It introduces neural networks as an alternative approach to the existing methods. The relationship between the number of nodes used and the accuracy of the neural network in modelling the cost flow is closely examined. From this research an optimal solution is proposed for the case and a prototype system is developed. The results of the investigation of the number of nodes used and testing of the prototype neural network for sample cases are presented and discussed.