【作者】 杜忠晓；

【导师】 王洪礼；

【作者基本信息】 天津大学 ， 管理科学与工程， 2009， 博士

【摘要】 建筑业的快速发展拉动了国民经济的持续增长,同时也为勘察设计企业提供了广阔的发展空间,但同时也面临着市场及管理等多方面的挑战。为了推动企业发展,既要加强对企业的效率评价,又要加强市场监管。为此,分别采用基于非参数法的数据包络分析(DEA)法和基于参数法的随机前沿面分析(SFA)模型对企业效率进行了有效评价,并开发出天津市勘察设计市场监管信息系统,具体如下:首先用DEA方法,将选取的45家典型的勘察设计企业作为评价对象,分别采用多输入单输出和多输入多输出指标,对企业的CRS技术效率、VRS技术效率和规模效率进行了评价。结果表明,有不少企业还有较大的效率提升空间。为了充分考虑到企业经营中的不可控随机因素(如气候、政策等)的影响,还应用SFA模型,对45家勘察设计企业的效率进行了评价。结果表明,两种方法都能在一定程度上反映企业的综合绩效。但SFA模型由于引入了随机误差,比DEA方法准确性高,可适用于“前沿生产函数”易于确立的情况;而DEA方法实用性强,且可用于多输出情况。因此,应针对具体情况进行方法的选择。在综合绩效影响因素的定量研究中,两种方法的联合运用更为重要。在企业效率评价中,最重要的输出指标就是设计面积,因此,在分析天津市1979-2007年城市房屋建筑(设计)面积数据特征的基础上,还采用随机梯度回归(SGR)法,对未来3年的天津城市建筑(设计)面积进行了预测。针对天津市勘察设计市场监管的特点,即设计市场管理(设计企业和设计人员的资质管理)与质量管理(施工图审查)同属一个部门,因此开发出了独具特色的、可将施工图审查中发现的设计错误自动记录到相应的企业和设计人员上(便于对其考核),同时具备市场管理和质量管理功能的天津市勘察设计市场监管系统。该系统采用多层体系结构和基于XML的ASP.NET技术,用B/S模式构建,已实际用于天津市勘察设计市场,既加强了市场监管,又提高了审图质量与速度,取得了良好的经济效益和社会效益。更多还原

【Abstract】 The rocketing growing of architectural industry stimulates the development of the national economy and offers a good opportunity for the development of architectural designing profession. But at the same time there are big challenges come from market and management. In order to promote enterprise development, it is necessary to strengthen the evaluation of the efficiency of enterprises and market supervision. This thesis evaluated the efficiency of enterprises with non-parameter data envelopment analysis (DEA) method and parameter stochastic frontier approach (SFA) method, and then designed and developed a market supervision information system.Based on the DEA method, the efficiencies of the 45 typical architectural designing enterprises were evaluated. The CRS technological efficiencies, VRS technological efficiencies and scale efficiencies were evaluated with multi-input multi-output and multi-input single-output indicators. Results showed that lot of enterprises had the ability to increase their efficiencies.In order to fully take into account the impact of uncontrollable random factors (such as climate, policies, etc), with the SFA model based parameters, the efficiencies of the 45 typical architectural designing enterprises were evaluated. Results showed that both methods can reflect the performance of enterprises to some extent. Due to the introduction of random error, SFA model has a higher accuracy than the DEA method, and is suitable in the situation of "forefront of the production function" is easy to establish; DEA method can be used for multi-output situation. Therefore, the situation should be specific for the method to use. It was important that both of these two methods should be used in the quantitative research of the efficiency evaluation.The most important output target is the designing area in the evaluating the efficiency of the enterprise. So under analyzing the characteristics of the Tianjin floor space constructed buildings, using stochastic gradient boosting of regression method, we predicted the floor space of the under constructed and finished buildings in 3 years based the historical data from 1979 to 2007.Considering the characteristics of Tianjin design market such as designed market management (design enterprises and designers of quality management) and quality management (review of construction plans) belong to one department, we built up the Tianjin architectural survey and design market supervision system. This system can record automatically the design errors in the corresponding enterprises and designers (to facilitate their assessments), and also has the market management and quality management function. The system used multi-tier architecture, B/S Mode, the XML-based ASP.NET technology, and not only strengthened the market supervision but improved the quality and speed of the trial plan. It has achieved good economic and social benefits.更多还原