机器学习在建设项目成本估算中的应用综述及展望

Review and prospects of the application of machine learning in construction project cost estimation

摘要:

成本被视为项目前期可行性研究和决策的主要标准之一。机器学习可以有效解决建设项目成本估算早期可用信息少的问题,故有越来越多的学者将机器学习应用到建设项目成本估算中来。因此,综述了多元线性回归(MLR)、支持向量机(SVM)、人工神经网络(ANN)、案例推理(CBR)和集成学习模型等五类机器学习方法在建设项目成本估算中的应用。本文发现机器学习有效提高了建筑项目成本估算的准确性和稳定性,并进一步对未来相关研究的发展进行了深入分析和展望。

Abstract:

Cost is considered one of the main criteria for pre-project feasibility studies and decision-making. Machine learning can effectively solve the problem of little available information in the early stage of construction project cost estimation. Therefore, more and more scholars are applying machine learning to construction project cost estimation. Therefore, the application of five types of machine learning methods in construction project cost estimation, including multiple linear regression (MLR), support vector machine (SVM), artificial neural network (ANN), case-based reasoning (CBR) and ensemble learning model, is reviewed. This article finds that machine learning can effectively improve the accuracy and stability of construction project cost estimation, and further conducts an in-depth analysis and outlook on the development of related research in the future.