【作者】 唐正娟；

【导师】 张涑贤；

【作者基本信息】 西安建筑科技大学 ， 管理科学与工程， 2010， 硕士

【摘要】 目前,我国正在进行很大规模的基本建设。而建筑业又是我国的支柱产业之一,是我国国民经济的重要组成部分。但是,由于建筑业独一无二的特点,使施工现场发生事故的可能性很高,这对国民经济和人们的生命和财产安全造成很大的损失。因此,我们必须对安全给予很高的重视,将安全管理工作作为工程项目管理的一个重要内容来看待。论文首先从安全管理研究入手介绍安全及安全评价的含义,进而对我国的建筑施工现场安全评价的特点以及我国建筑安全施工现场存在的问题进行了论述。在阅读了大量文献资料的基础上,对我国目前建筑安全的研究内容和研究方法进行了分类和总结,并提出了基于BP神经网络的建筑施工现场安全评价模型。本文首先对影响建筑施工现场安全状况的众多因素进行了分析,主要是从人的因素、建筑材料及设备因素、技术因素、管理因素和环境因素五个方面进行细化。结合我国建设部颁布的《施工企业安全生产评价标准》(JGJ/T77-2003)中的各项指标,建立了一套更新的建筑安全评价指标体系。包含了24个影响建筑施工现场安全的主要指标。在建立好的建筑施工现场安全评价指标的基础上,结合人工神经网络的理论建立了基于BP神经网络的建筑施工现场安全评价模型。本文通过用用专家评分法来获取BP神经网络的样本数据。通过评分表的方式向几位专家调查打分,根据给定的每位专家的权重和专家所打的分数来计算施工现场的安全程度分数。通过统计后得到了15个建筑施工项目的评价指标体系指标值作为训练BP神经网络的样本数据。通过用专家打分法产生的样本数据对网络进行训练,结果证明可以发现,期望输出与实际输出的误差很小,接近一致,这充分说明运用BP神经网络模型建立的建筑施工现场安全评价模型是有效的,可以使用这个模型对建筑施工现场安全进行评价的实证研究。更多还原

【Abstract】 At present, China is undergoing a large-scale construction in its history. And the construction industry is one of the anchors in China. Because of the characteristics of construction, there is always full of the potential safety problems in the construction locale. All these kinds of factors may lead to accidents some time, and threaten the lives safety of construction workers seriously, so the safety management becomes an important part of the project management.First of all, the papers talked about the concept of security and safety assessment, and then discussed the feature and the condition of the construction of China. After reading a lot of literature, the writer classified and summarized the current research content and methods of construction safety management in China, then a construction safety assessment model based on BP Neural Network was put forward.Many factors influent the construction site safety was talked about firstly. The factors include human, material and mechanical equipment, technique, management and the environment of the construction site. The author set up a new construction safety assessment criteria system, which referred the "Safety Evaluation Criteria in Construction Enterprises" (JGJ/T77-2003) promulgated by China’s Ministry of Construction. It includes 24 indicators in the system. On the base of the indicator system, the writer built a model based on the BP Neural Network. The model of BP network will be training by many training samples which came from the expert scores. By given the weight, many experts graded the construction site safety condition, and 15 training samples were got.Using the sample data generated by the method of expert scoring to train the network, the results show that:there is small error between the desired output and actual output. So that the model of site safety assessment of construction using the model of BP neural network is effective, and it can be used for the empirical study of construction site safety assessment.更多还原