基于SVM-EVM的EPC项目施工阶段进度—成本预测研究

发布时间：2018-02-11 01:00

本文关键词： EPC项目施工阶段挣值法进度—成本预测支持向量机模型　出处：《河北工程大学》2017年硕士论文　论文类型：学位论文

【摘要】：在目前国际市场新形势下,EPC总承包模式的实施逐步成为国内建筑行业达到国际标准的重要途径。但是,国内总承包商在对项目成本和进度控制上还是缺少理论和实践经验,整体水平和综合实力与国际先进水平仍有很大差异。在市场竞争力存在很大的不足的情况下,对EPC项目进度和成本集成管理直接决定了建设项目的成败。因此有必要对EPC项目进度和成本进行整体管理预测,提高项目的经济和社会效益。本文围绕EPC项目进度和成本管理预测进行研究。首先具体分析施工阶段进度和成本的关系以及相互影响的机理,介绍集成控制的基本理论和流程。通过在现有的建设工程施工阶段的成本控制基础上,提出了科学有效的控制措施和方法,同时将进度—成本预测应用于施工阶段成本控制中。但是,在实际建筑工程进度—成本预测的研究中发现,传统的预测方法存在不严谨以及预测精度不足等问题。所以本文先用EVM进行数据预处理,得到绩效指数,进而用支持向量回归模型来进行项目施工阶段进度—成本预测,为了得到更好地预测效果,用粒子群进行参数优化来对预测点以后紧邻工程的进度、成本绩效指数进行预测。最后将此预测模型应用于EPC项目施工阶段进度—成本预测案例中,得到预测结果,将原始数据与预测数据进行对比分析,从而验证该模型的可操作性和有效性,提高工程项目施工阶段进度—成本的管理水平。
[Abstract]:Under the new situation in the international market, the implementation of EPC general contracting mode has gradually become an important way for the domestic construction industry to meet the international standards. However, the domestic general contractor still lacks theoretical and practical experience in controlling the project cost and schedule. The overall level and comprehensive strength are still very different from the international advanced level. The integrated management of EPC project schedule and cost directly determines the success or failure of the construction project. Therefore, it is necessary to forecast the progress and cost of the EPC project as a whole. To improve the economic and social benefits of the project. This paper focuses on the prediction of EPC project schedule and cost management. Firstly, the relationship between schedule and cost in construction phase and the mechanism of mutual influence are analyzed. The basic theory and flow chart of integrated control are introduced. Based on the cost control in the construction stage of the existing construction project, the scientific and effective control measures and methods are put forward. At the same time, the schedule cost forecast is applied to the cost control in the construction stage. However, in the research of the actual construction progress cost forecast, it is found that, There are some problems in the traditional prediction methods, such as unrigorous prediction method and insufficient prediction precision, so this paper first uses EVM to preprocess data and get the performance index, and then uses support vector regression model to predict the schedule and cost of the construction phase of the project. In order to obtain better prediction effect, particle swarm optimization is used to predict the schedule and cost performance index of the project next to the prediction point. Finally, the prediction model is applied to the case of EPC project construction phase scheduling and cost forecasting. The prediction results are obtained by comparing the original data with the predicted data to verify the maneuverability and effectiveness of the model and to improve the management level of the schedule and cost in the construction phase of the project.
【学位授予单位】：河北工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU723.3

【参考文献】