Please use this identifier to cite or link to this item:
|Title||Parametric Cost Estimation of Road Projects Using Artificial Neural Networks|
Early stage cost estimate plays a significant role in any initial road projects decisions; despite the project scope has not yet been finalized. Major problems faced are lack of preliminary information, database of road works costs and appropriate cost estimation methods. Therefore, it is important to find a way to estimate the roads cost in a short time with acceptable accuracy. The main objective of this research is to develop artificial neural networks model to estimate the cost at early stage of road projects in Gaza strip using parametric techniques and reduce the percentage error of estimation. To achieve this there is a need to identify the factors that affect the cost of road projects that can be available at early stage. Three types of networks were used to build the models: linear regression (LR) as a traditional method for estimating, and the two most common kinds of feed forward patterns, which are multilayer perceptron (MLP) and general feedforward (GFF). Topologies and architectures of ANN were adopted after several trials. The total accuracy performances of the best model of each type are 90.3%, 93.72% 94.5% for LR, MLP and GFF model respectively. A visual basic interface was developed to operate GFF model and facilitate data entry. The GFF was the best model that had mean absolute percentage error 4.98%, 4.57% and 6.09 for training, cross validation and test sets respectively. Sensitivity analysis was done for GFF model, which showed that the pavement area parameter had the greatest effect on the budget output but project scope had a low impact. This research was achieved the ability to estimate the cost of road projects at early stage and reducing the error rate to reach 5.5%. ANN is well suited to model complex problems where the relationship between the model variables is unknown so take advantage of the ANN capabilities in the construction management.
|Publisher||الجامعة الإسلامية - غزة|
|Files in this item|