Cost estimation has its proven importance as one of essential factors for project success. The aim of this research is to predict the early project cost using neural network. Early project cost represents a key compon...Cost estimation has its proven importance as one of essential factors for project success. The aim of this research is to predict the early project cost using neural network. Early project cost represents a key component in business unit decisions. The most important factors influencing on the parametric cost estimation in construction building projects in Gaza Strip were defined and investigated. A questionnaire survey and relative index ranking technique were used to conclude the most important factors. Fourteen most effective factors were identified. One hundred and six case studies from real executed construction project in Gaza Strip were collected for training and testing the model. The cases were prepared to be used in cost estimate neural networks model. Eighty percent of case studies were used to train and test the model. The remaining 20% was used for model verification. The results revealed the ability to the model to predict cost estimate to an acceptable degree of accuracy. The minimum squares error with 0.005 in training stage and 0.021 in testing stage were recorded.展开更多
文摘Cost estimation has its proven importance as one of essential factors for project success. The aim of this research is to predict the early project cost using neural network. Early project cost represents a key component in business unit decisions. The most important factors influencing on the parametric cost estimation in construction building projects in Gaza Strip were defined and investigated. A questionnaire survey and relative index ranking technique were used to conclude the most important factors. Fourteen most effective factors were identified. One hundred and six case studies from real executed construction project in Gaza Strip were collected for training and testing the model. The cases were prepared to be used in cost estimate neural networks model. Eighty percent of case studies were used to train and test the model. The remaining 20% was used for model verification. The results revealed the ability to the model to predict cost estimate to an acceptable degree of accuracy. The minimum squares error with 0.005 in training stage and 0.021 in testing stage were recorded.
