重大工程施工串谋监管动态性分析与策略优化

发布时间：2018-01-02 10:19

本文关键词：重大工程施工串谋监管动态性分析与策略优化　出处：《华中科技大学》2015年硕士论文　论文类型：学位论文

【摘要】：政府代建制下的公共基础设施工程项目中,针对监理和承包商施工现场串谋行为的监管作为保障工程质量的重要一环在项目管理中尤为重要。目前基于传统博弈论分析工程串谋行为重在博弈均衡解的存在性分析,忽略实际串谋问题中行为人的有限理性特征引起博弈决策的不确定性。长期过程中串谋决策的动态反复性和复杂多变性会导致政府监管部门对串谋行为的预测和控制策略失效。鉴于此,本文尝试将演化博弈以及系统动力学动态仿真软件相结合,研究有限理性条件下政府工程施工串谋监管的动态性分析以及监管策略优化问题。本文首先对政府工程代建流程以及串谋类型进行分析,确定以施工过程中监理承包商组成的串谋体为监管打击对象。在引入前景理论刻画监理风险态度的条件下,分析博弈参与方利益关系并建立博弈矩阵求解Nash均衡。进一步通过串谋监管混合战略博弈系统动力学模型仿真揭示博弈Nash均衡决策的动态反复性,并指出一般控制策略难以遏制决策波动。在此基础上,建立有限理性条件下政府串谋监管的演化博弈模型,通过雅克比矩阵分析判定演化稳定策略ESS。然后运用演化博弈系统动力学建模仿真对稳定策略进行验证。针对不存在ESS的情况,提出双重惩罚机制以及动态惩罚机制的优化策略抑制串谋决策的不稳定波动。最后,根据仿真分析对政府串谋监管提出实际指导建议,如差异化监管、科学惩罚机制、成本控制、健全行业信誉体系等。本文突破性的将前景理论和演化博弈动态性研究以及系统动力学仿真相结合分析工程施工串谋监管问题,为未来的研究在理论和技术工具上提供了很好的借鉴方向。
[Abstract]:On behalf of the government under the construction of public infrastructure projects, the construction site supervision and contractor collusion behavior supervision as an important part of ensuring engineering quality in project management is particularly important. At present, based on the traditional game theory analysis on engineering collusion behavior game equilibrium analysis of the existence of solutions, ignoring the limited rational of the actual behavior of collusion problem in the game caused by the uncertainty of decision making. The dynamic recurrent collusion process and long-term decision-making complexity will lead to government failure prediction and control strategy of the collusion behavior. In view of this, this paper tries to apply evolutionary game theory and system dynamics simulation software combined with construction supervision of the government conspiracy to dynamic analysis and engineering study on the supervision strategy optimization problem under limited rationality conditions. Firstly, the government project construction process and conspiracy type Analysis of body composition to determine the collusion in the process of construction supervision contractor for the supervision target. The introduction of prospect theory describe the risk attitude of the supervision conditions, analysis of game interests and sets up a game matrix to solve the Nash equilibrium. Further dynamic recurrent mixed supervision through collusion strategy game system dynamics model simulation reveals the game equilibrium decision Nash the general control strategy, and pointed out that the decision is difficult to curb volatility. On this basis, the evolution game model of government supervision to establish collusion under limited rationality, and then determine the evolutionary stable strategy ESS. using evolutionary game system dynamics modeling and simulation to verify the stable strategy by Jacobian matrix analysis. According to ESS, the proposed optimization strategy of double the punishment mechanism and punishment mechanism of dynamic decision making unstable suppression conspiracy. Finally, root According to the simulation analysis puts forward the practical suggestions for the government to guide the supervision, such as the difference of scientific supervision, punishment mechanism, cost control, and improve the credibility of the industry system. This article breakthrough the prospect theory and evolutionary game dynamics research and system dynamics simulation combined analysis of collusion problems in engineering construction supervision, for future research in theory and technology the tool provides a good direction.

【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TU71

【相似文献】