基于数据包络分析方法的地下道路交通工程设计方案评价

Evaluation on underground road transportation engineering design project based on data envelopment analysis

为了评价地下道路交通工程设计方案的优劣性,建立了以安全、技术、经济、运行效率和环境为导向的地下道路交通工程设计方案评价体系;引入数据包络分析(data envelopment analysis,DEA)方法,分析其对地下道路交通工程设计评价的适应性;根据DEA方法对输入输出指标同向性的要求,选取地下道路的冲突点个数、建设投资费用、交通延误作为评价的输入指标,将设计通行能力和照度作为评价的输出指标,并充分考虑指标与评价对象的关联性和易量化性;最后以西安市创业新大陆地下遂道项目为例,用选取的指标对DEA方法进行验证。利用DEA方法模型中的规模收益可变的BCC模型(由Banker、Charnes和Cooper提出),得到10个地下道路交通工程设计方案的综合效率评价值,进而对这10个设计方案的技术效率和规模效率进行分析,指出各设计方案DEA无效的原因,并给出部分方案的改进方向和建议。研究结果表明:相较于规模收益不变的CCR模型(由Charnes、Cooper和Rhodes提出),规模收益可变的BCC模型更适合评价地下道路交通工程设计方案,得到地下道路不同工程设计方案的综合效率评价值;当有多个设计方案的综合效率值为1时,能给出这些方案的优劣排序,还可计算出各DEA无效方案的输入冗余率和输出亏空率,指出各设计方案的改进方向和步长,避免投入产出要素的过剩或不足。

In order to evaluate the performance of underground road transportation engineering design projects, an evaluation hierarchy of underground road transportation engineering design project was established based on oriented security, technology, economy, efficiency of operation and environment. Data envelopment analysis （DEA） method was introduced to analyze the adaptation of underground road transportation engineering design evaluation. According tohomonymous provision of DEA to input and output indexes, number of conflict point, construction costs and traffic delay were chosen as input indexes, and design traffic capacity and illuminance were chosen as output indexes, which took full account of the relevance between indexes and evaluation objects, and easy to be quantified. Finally, by taking the New World Underground Tunnel Project in Xi＇an as an example, the DEA method was verified by the selected indexes, using BCC （propose by Banker, Charnes and Cooper） model, a special DEA method with variable returns to scale, was used to get the overall efficiency of the 10 underground road transportation engineering design projects, and then the technical efficiency and scale efficiency of those 10 design projects were analyzed and the cause of DEA invalid in each project was also revealed, so that the improvement direction and advice of some projects could be pointed out. The results show that compared with CCR （proposed by Charnes, Cooper and Rhodes） model whose returns to scale is constant, BCC model has a stronger adaptation to evaluate underground road transportation engineering design projects, and get the overall efficiency of different underground road transportation engineering design projects. When the overall efficiency of multiple design projects is 1, the method can not only obtain the relative merits of those projects, but also calculate input redundancy rate and output deficiency rate of each DEA invalid project to point out the improvement direction and step size, avoiding deficiency or surplus of input and output factors as well as wasting of resources. 3 tabs, 2 figs, 16 refs.