基于粗糙集_神经网络的工程项目质量风险评价研究

发布时间：2018-02-10 05:40

本文关键词： 工程项目质量风险评价粗糙集神经网络指标体系　出处：《江西理工大学》2013年硕士论文　论文类型：学位论文

【摘要】：近年来，建筑业在我国得到了快速发展，建筑市场的规模也在逐渐地扩大，但由于工程项目自身具有规模大、周期长、投资高等特性，使得工程项目的质量问题越来越受到大家的重视。一旦工程项目的质量出现问题，不仅会影响到整个工程项目的交付和使用，造成经济利益的损失，如果严重的话还会影响到国计民生甚至整个建筑业的发展。目前，工程项目的质量问题已经成为人们关注的焦点，人们希望通过技术控制和科学管理使得工程项目质量达到预期目标，而工程项目质量风险评价是工程项目质量风险管理中最有效的管理方法，是发现质量问题、找出问题原因和保证工程项目质量的重要依据。因此，本文对工程项目质量风险评价展开了理论和实证研究，希望能够给我国的工程项目质量风险管理提供一定的理论依据。本文首先从工程项目质量风险管理入手，进而对工程项目质量风险管理的相关概念和特点展开了详细的论述。然后分析了工程项目质量风险管理的全过程，逐一对传统的风险识别和风险评价方法进行了相关介绍，并结合工程项目自身的特点，提出了基于粗糙集_神经网络的工程项目质量风险评价方法，同时论证了该评价方法的可行性。接下来根据工程项目施工阶段的实际情况，运用相关理论和方法识别出工程项目施工阶段的质量风险因素，构建出工程项目质量风险评价指标体系。最后对粗糙集_神经网络风险评价模型进行了实证性分析。首先利用粗糙集理论对收集到的工地样本数据进行处理，然后将约简后的评价指标作为输入端建立了工程项目质量风险的神经网络评价模型，，并利用MATLAB7.0软件工具对BP神经网络进行训练和检验。通过实证分析的结果可以看出该模型的可操作性较好，评价的效果不错，具有一定的应用价值。
[Abstract]:In recent years, the construction industry has been developing rapidly in our country, and the scale of the construction market is also gradually expanding. However, because of the large scale, long period, high investment and so on, the engineering project itself has the characteristics of large scale, long period, high investment and so on. The quality problem of the engineering project is paid more and more attention to. Once the quality of the project has a problem, it will not only affect the delivery and use of the whole project, but also cause the loss of economic benefits. If it is serious, it will also affect the development of the national economy and the people's livelihood and even the entire construction industry. At present, the quality of engineering projects has become the focus of attention. It is hoped that the project quality can reach the expected goal through technical control and scientific management, and the engineering project quality risk assessment is the most effective management method in the engineering project quality risk management, which is to find the quality problem. To find out the cause of the problem and the important basis to guarantee the quality of engineering project, this paper has carried out theoretical and empirical research on quality risk assessment of engineering project, hoping to provide some theoretical basis for quality risk management of engineering project in our country. This paper begins with the quality risk management of engineering projects, and then discusses in detail the related concepts and characteristics of quality risk management of engineering projects, and then analyzes the whole process of quality risk management of engineering projects. This paper introduces the traditional risk identification and risk assessment methods one by one, and puts forward the quality risk evaluation method based on rough set _ neural network, combining with the characteristics of engineering project itself. At the same time, the feasibility of the evaluation method is demonstrated. Then, according to the actual situation of the construction phase of the project, the quality risk factors of the construction phase of the project are identified by using the relevant theories and methods. The quality risk evaluation index system of engineering project is constructed. Finally, the rough set _ neural network risk evaluation model is analyzed empirically. Firstly, the collected site sample data are processed by rough set theory. Then the reduced evaluation index is used as the input to establish the neural network evaluation model of engineering project quality risk. The MATLAB7.0 software is used to train and test BP neural network. The results of empirical analysis show that the model has good operability, good evaluation effect and certain application value.
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU712.3

【参考文献】