Artificial neural networks for improving cost model development process

In order to cope the changes occurring in U.K. industry within the next decade, it is expected that the quantity, type, accuracy and complexity of cost models will need to be greatly increased. The objective of this study is therefore to improve the speed and effectiveness with which cost models can be developed and applied. This current work is concerned with identifying and examining current mathematical modelling techniques used to establish CER's and determining their relative advantages and limitations in terms of their effects on the overall cost model development process. Experimental results are presented that indicate the advantages of using artificial neural networks (ANN) models when compared with traditional regression based techniques.