The cost prediction of an overhead line reconstruction project represents an important part of management and control of the project, and to improve the prediction accuracy, a combinational forecasting method is presented. An overhead line reconstruction project is first decomposed into several subprojects depending on the characteristics of the project, and the cost prediction could be made first for each subproject and then integrated. The principal component analysis and expertise of domain experts are combined to get the key factors of the subprojects. The genetic algorithm based support vector machine and extreme gradient boosting algorithm are then used for cost prediction respectively. The Shapley value theory in game theory is next employed to determine the weights of combinational forecasting so as to attain an appropriate combinational forecasting model. Finally, an actual overhead line reconstruction project is employed to demonstrate the presented combinatorial forecasting model, and more accurate prediction result is attained, compared with those attained by the genetic algorithm based support vector machine and extreme gradient boosting system independently.