【作者】 杨波；

【导师】 余建星；

【作者基本信息】 天津大学 ， 结构工程， 2007， 博士

【摘要】 本研究以大型集装箱港口的建设作为主要的的研究对象,以其中的决策和建设环节目前相对薄弱的理论环节-影响决策的数量化因素,集装箱运量的预测,融资、建设过程的风险分析作为核心研究内容,形成了本论文的两个主要部分:本论文总结了集装箱吞吐量预测的常用方法,并且根据中长期预测的要求对于各种方法的使用以及相关参数的选择进行了比较研究。然后,根据模型的机理对于方法的可靠性进行了评价,最终给出了综合权重矩阵。多因素动态生成系数法是目前集装箱吞吐量预测方法中最为成熟、应用最为广泛的预测方法。该办法首先确定外贸年增长率预测外贸量,然后通过生成系数得到集装箱吞吐量预测值。然而,在进行中长期集装箱吞吐量预测时,未来外贸年增长率的确定经常缺乏依据,导致长期预测结果精度较差。本论文通过利用逻辑斯蒂增长模型进行中长期经济增长预测,再利用经济增长与外贸量的回归分析进行外贸量预测来解决这一问题,最终以天津港2005-2020年集装箱吞吐量预测为案例进行了预测。从原理和结果上看,逻辑斯蒂增长模型是更为适宜和先进的方法,可以用来改进多因素动态生成系数法,提高中长期集装箱预测的精度。风险管理是近二十年发展起来的一门新兴学科,涵盖了多个学科的综合运用,近年来开始广泛应用于各类工程项目。本文针对集装箱港口建设的融资及建设风险进行了深入的研究,目的是通过对港口建设项目的融资及建设风险进行分析,研究更有效的风险分析方法。本论文首先分析了融资过程中风险管理的一般过程;接着基于融资者对不同风险所持有的风险偏好,建立风险度量函数模型,并用此来定量计算风险值大小;然后通过案例,具体分析研究了东疆港区中项目融资的诸多风险表现形式,并计算出其中风险值,同时基于风险大小提出防范风险的具体措施。接着重点研究了建设过程中存在的主要风险。并利用主客观结合的赋权法和为了突出主要因素所采用的变权方法来计算建设风险。通过实例演算,分别验证了两种方法的有效性。通过与常规方法对比,可以看到两种方法的优越性。本文为了简化计算,使计算过程便于理解,过程采用矩阵计算,结果利用向量表示,更加一目了然。本研究在基础理论分析研究的基础上,以天津港东疆港区开发作为范例,分别对其运量预测,融资、建设风险进行了系统的分析和探讨,从而为港口建设决策提供了重要的定量分析参考。更多还原

【Abstract】 This paper summaries methods for container handling capacity prediction and then gives suggestions on method selection and related parameters’setting according to the characters of middle/long term prediction. Then, the reliability of each method is assessed based on its mechanism to produce the synthesis weight matrix. At last, this paper uses the Tianjin Port as a case to conduct middle/long term container handling capacity prediction.Multi-factor dynamic product coefficients method is the most mature and widely-used one in all available container handling capacity prediction models. After predicting foreign trade value, predicted container handling capacity can be got by using product coefficients. However, the very first step, setting future foreign exchanging growth rate, is a step which usually lacks solid logical base, leading to low precision in the middle/long term prediction. This paper adopts logistic model to make middle/long term GDP, then solves this logic-lacking problem by using regression model between economic volume and foreign trade value. The middle/long term container handling capacity for the Tianjin Port is conducted as a case. The result and rationales show that logistic model is a more appropriate and advanced method to modify the multi-factor dynamic product coefficients and improve middle/long term prediction’s precision.Risk management is a new subject, which was developed in the past twenty years, and starts to be comprehensively applied to all kinds of projects recently. In this paper, thorough study on financing and construction risk of port development was conducted. Purpose of this study was to find out the more effective risk analysis method for container port development.Firstly the common process of risk management was concisely analyzed; then based on the investors’different risk preference, a math’s model was set up to measure risk; finally by providing an example, various material risks included in this example was studied up on and the numerical value of them was calculated, and then some measures to avoid risks was offered in virtue of the numerical value of the risks.Then it put the key research on the combination weighting approach as well as the variable weighting approach to emphases on the dominant factor. The importance of hazards in the bottom layer can be appraised objectively by the combination weighting approach while the importance can be emphasized and using the variable weighting approach can enhance the authenticity. This paper checks the foregoing two methods by examples calculation and method comparison. In order to simplify calculation and make the course easily understood, matrix computation was applied in the process and vector was applied as a consequence,being clear at a glance all through this paper.Financing and construction process are two of the most important steps in the container terminal development. Based on the theory study of this paper ,The author use the principle to Tiajin Port east area container terminal financing and construction risk analysis.更多还原