Cost Estimation and Sensitivity Analysis on Cost Factors: A Case Study on Taylor Kriging, Regression and Artificial Neural Networks

Abstract

This article applies Taylor kriging (TK) to cost estimation. The partial differentiation equation of TK is developed and used to assist in sensitivity analysis on cost factors. The capabilities of cost estimation and sensitivity analysis of TK are compared with those of regression and an artificial neural network (ANN) through application in a case study. The results show that TK can provide more accurate cost estimates than those of regression but worse than those of ANN, and both TK and regression are able to effectively find sensitive cost factors.