考虑风险不确定性与动态性的项目工期估算

发布时间：2019-03-26 09:09

【摘要】：在建筑工程项目众多的风险之中,工期风险居于核心地位,因为工期的延误将使项目各利益相关方的效用均下降,使项目的收益低下。特别是业主,假如项目不能按照预先计划顺利建成投产,那么项目存在的必要性就会大打折扣,工期风险带来的损失不仅包括银行贷款利息的增加、投资成本增加、现金流无法按计划实现,还可能导致商机的丧失。通常意义下的工程项目进度计划是以关键线路法CPM(Critical Path Method)为基础的,它的基本假定是网络进度计划中每项活动的持续时间是确定的。而实际工程中,由于政治、经济、气象、水文、施工环境等不确定性因素的影响,导致了工程项目施工中工序活动持续时间的不确定性,因而就出现了如何计算这种网络进度的时间参数以及工期能否按计划实现的问题。本文在研究经典PERT网络计划和蒙特卡罗模拟的基础上,运用灰色系统理论和马尔科夫决策两大不确定性研究方法,分别建立了灰色网络计划模型和考虑动态风险的马尔科夫模型。第一,本文对比了经典PERT网络计划和蒙特卡罗模拟结果的差异,通过统计各工序出现在关键路径上的次数产生对下面问题的思考：关键工序的衡量标准是什么?不确定型网络计划模型中关键路径该如何定义?接着以数学推理的形式解释了出现上述差异的原因：忽略了非关键路径的影响。第二,本文建立了区间网络计划模型,通过引入灰数排序规则构建了完整的区间网络计划时间参数计算规则。为了减弱灰度放大效应的干扰,对超出合理区间的部分进行逐一修正。第三,本文在区间网络计划模型的基础上,进一步探索了带三角白化权函数的网络计划模型,其主要目的是引入取值分布情况以更准确地反映灰信息。第四,对地震和罢工两大不可抗力影响下的工期估算进行了必要的探索,得到了符合动态决策的工期估算方法。本文尚不能构建不确定型网络计划的整个理论体系和框架,其部分章节的研究结果依然存在继续改进的可能性。论文的结尾总结了几大值得继续探索的研究方向,希望能够给未来研究提供一定的思想借鉴。
[Abstract]:Among the many risks in the construction project, the construction project risk occupies the core position, because the delay of the construction period will reduce the utility of all stakeholders of the project, and make the project benefit low. In particular, owners, if the project cannot be successfully completed and put into operation according to the pre-planned plan, then the necessity of the project will be greatly reduced, and the loss brought about by the time limit risk will not only include the increase in interest on bank loans, but also the increase in investment cost. Cash flow cannot be realized as planned, and it may lead to the loss of business opportunities. In general, the project schedule is based on the critical path method (CPM (Critical Path Method), and its basic assumption is that the duration of each activity in the network schedule is determined. In the actual project, due to the influence of the uncertain factors such as politics, economy, meteorology, hydrology, construction environment and so on, it leads to the uncertainty of the duration of the process activities in the construction of the engineering project. Therefore, how to calculate the time parameters of the network progress and whether the time limit can be realized according to the plan arises. On the basis of studying the classical PERT network plan and Monte Carlo simulation, the grey system theory and Markov decision-making are used to study the uncertainty. Grey network planning model and Markov model considering dynamic risk are established respectively. First, this paper compares the difference between the classical PERT network plan and the Monte Carlo simulation results. By counting the number of times that each process appears on the critical path, this paper gives some thoughts on the following questions: what is the measure standard of the key process? How do critical paths in an uncertain network planning model be defined? Then the reason of the difference is explained in the form of mathematical reasoning: the influence of non-critical path is ignored. Secondly, this paper sets up the interval network planning model, and constructs the complete interval network planning time parameter calculation rule by introducing the grey number ranking rule. In order to reduce the interference of the magnification effect of gray scale, the parts out of the reasonable range are modified one by one. Thirdly, on the basis of the interval network planning model, this paper further explores the network planning model with triangular whitening weight function, the main purpose of which is to introduce the value distribution to reflect the grey information more accurately. Fourthly, it is necessary to explore the time limit estimation under the influence of earthquake and strike force majeure, and obtain the time limit estimation method in accordance with dynamic decision-making. This paper has not been able to construct the whole theoretical system and framework of uncertain network planning, and the results of some chapters still have the possibility of further improvement. At the end of the paper, several research directions worth further exploration are summarized, hoping to provide some ideas for future research.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU72

【参考文献】