Abstract:

The cost estimation of the utility tunnel is commonly characterized by multiple influencing factors, small sample and nonlinearity; and traditional cost estimation method is not effective. As a result, a cost estimation model for utility tunnel based on genetic algorithm (GA) and support vector machine (SVM) is proposed. Firstly, the specific characteristics of the utility tunnel are analyzed; and then the nonlinear mapping relationship between engineering cost and influencing factors is established by SVM. Finally, the parameterized optimization for SVM and the cost estimation of the utility tunnel are carried out by GA. The model is analyzed and verified by 16 existing utility tunnels; and the sample test results show that the estimation error can be controlled within 10%, so as to verify the feasibility of the model.

Key words: utility tunnel, support vector machine (SVM), genetic algorithm (GA), cost estimation