机器学习在建设项目成本估算中的应用综述及展望  

A Review and Prospect of the Application of Machine Learning in Construction Project Cost Estimation

作　　者：李敬业 翟武娟 宁延 Li Jingye;Zhai Wujuan;Ning Yan(School of Management and Engineering,Nanjing University,Nanjing 210093,China)

机构地区：[1]南京大学工程管理学院,江苏南京210093

出　　处：《科技管理研究》2024年第20期166-173,共8页Science and Technology Management Research

基　　金：国家自然科学基金面上项目“全生命周期视角下基础设施项目价值共创研究:目标、行为及治理”(72271118)。

摘　　要：成本在项目前期可行性研究和决策过程中占据关键地位,被视为主要标准之一。机器学习可以有效解决建设项目成本估算早期可用信息有限的问题,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。运用多元线性回归(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5类机器学习方法对建设项目成本估算中的应用情况进行了系统综述,研究结果发现,这些机器学习方法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,为投资决策者提供参考。Cost estimation plays a critical role in feasibility studies and decision-making processes during the early stages of construction projects.Given the limited availability of data in early cost estimations,machine learning(ML)techniques have emerged as effective tools to address these challenges.This paper systematically reviews the application of five major machine learning methods in construction cost estimation:Multiple Linear Regression(MLR),Support Vector Machines(SVM),Artificial Neural Networks(ANN),Case-Based Reasoning(CBR),and Ensemble Learning Models.Through an analysis of their performance and improvements,the findings reveal that these machine learning techniques significantly enhance both the accuracy and stability of cost predictions.Moreover,the paper offers insights into future research trends,focusing on refining algorithms to handle the inherent uncertainties in early project phases and improving the interpretability of the models.it aims to provide valuable guidance for project stakeholders and decision-makers to optimize their investment strategies and risk management in construction projects.

关 键 词：机器学习 建设项目 成本估算 研究综述 

分 类 号：TU723.3[建筑科学—建筑技术科学] F423.1[经济管理—产业经济] F28[文化科学] G301