【作者】 张小琳；

【导师】 朱明侠；

【作者基本信息】 对外经济贸易大学 ， 国际贸易学， 2014， 博士

【摘要】 石油是现代工业生产发展赖以运行的基本能源，被称之为现代工业的“血液”，在我国经济发展中占有举足轻重的地位。近年来，我国经济快速增长带动石油消费持续增加，我国已成为仅次于美国的世界石油消费第二大国。由于国内石油生产能力不能满足国内日益增加的石油需求，石油进口依存度持续攀升，我国已超过美国成为世界石油第一大净进口国，石油需求严重依赖外部供应。本文以石油贸易风险为研究对象，采用定性与定量相结合的方法，以国际贸易理论、风险管理理论、层次分析法理论、遗传算法理论等相关理论为基础，采用遗传算法修正的层次分析法（GA-AHP）对石油贸易风险影响因素进行分析，确定影响石油贸易的各风险因素权重，进行石油贸易风险评估。石油贸易风险影响因素错综复杂，本文首先在全面梳理国内外相关研究成果的基础上，向行业内专家进行咨询请教，对影响我国石油贸易的风险因素进行识别，构建了包含目标层、准则层、子准则层和方案层的多层次指标体系。采用系统化的、层次化的方法对石油贸易风险各因素进行评估。其次，由于层次分析法存在一致性问题，论文使用遗传算法进行了修正，并通过杂交变异对多个专家的判断矩阵进行处理，对我国石油贸易风险做出定量评估。最后，本文还使用遗传算法修正的层次分析法对管理我国石油贸易风险的重要战略对策进行综合评价优选。本文根据遗传算法修正的层次分析法得出以下主要结论：一是政治风险对石油贸易的影响权重最大，是石油贸易中面临的主要风险；油源风险次之；市场风险和运输风险对石油贸易风险的影响权重相对较小。二是对于防范石油贸易风险的五个基本方案而言，开展石油外交是国际石油贸易风险管理最有效的措施，以下依次是实施多元化策略、建立石油金融市场、采取合理的运输策略、建立国家石油储备。根据论文研究结果，本文建议我国应积极开展石油外交，与油源国保持友好的战略合作关系，实施多元化策略，分散油源集中风险。本文的贡献性主要体现在：一是采用定量方法综合分析石油贸易风险问题；二是在进行文献研究并与相关专家讨论，构建了风险多层次指标体系；三是将遗传算法与层次分析法相结合，对传统层次分析法中的一致性问题进行修正，构建了遗传算法修正的层次分析法（GA-AHP）。由于本人学术水平仍有待提高，对石油贸易的认识仍需深入，本论文还存在一定的局限性。由于AHP方法只考虑了因素间自下而上的影响关系，对于较复杂的网络型、反馈型问题，AHP方法甚至GA-AHP方法的有效性存在疑问，而现实情况中因素间的网络型关系或多或少地存在，如何处理网络型、反馈型的影响关系，将是作者未来需要研究的课题。更多还原

【Abstract】 Oil is the basic energy for the development of modern industrial production andis called “blood” of modern industry. Oil plays an important role in China’s economicdevelopment. In recent years, as the oil consumption continues to grow driven byrapid economic growth, China has become the world’s second largest consumer of oilafter the United States. As domestic oil production capacity cannot meet theincreasing domestic demand, the degree of dependence on the foreign oil has beenincreased greatly. China has surpassed the U.S. as the world’s largest net importer ofoil and China’s oil demand is heavily dependent on external supply.This paper takes China’s international oil trade as research object and theresearch is based on international trade theory, risk management theory, analysishierarchy process, and genetic algorithm theory and so on. This paper determines therisk weight of various risk factors on oil trade by using Genetic Algorithm RevisedAHP (GA-AHP), which is a combination of qualitative and quantitative method.The risk factors of international oil trade are intricate and complex. First of all,this paper identifies risk factors and constructs a multi-layer indicator hierarchy whichincludes the overall goal, the criteria layer, the sub-criteria layer and the decisionalternatives, by making comprehensive literature reviews related to this subject athome and abroad and by consulting oil industry experts. This paper uses a systematic,hierarchical approach to assess the risk factors of international oil trade. Secondly,because of the consistency issue of AHP, this paper uses genetic algorithm method torevise the judgment matrix consistency and process all experts’ judgment matrixthrough crossover and mutation operations. Finally, this paper makes comprehensiveevaluation on important strategic countermeasures of international oil trade riskmanagement by using GA-AHP.This paper draws several main conclusions based on GA-AHP. First of whichsays that, while market and transportation risks are relatively small, the primary riskin the international oil trade comes from political factors, which accounts for thelargest portion of the overall weights, followed by factors about oil sources. Second,our paper indicates that, among the5decision alternaltives, energy diplomacy is themost efficient measure to manage the risks of oil trade, followed by diversificationstrategy, setting up oil finance market, reasonable transportation strategy, andestablishment of National Petroleum Reserve system, respectively. According to the results of this paper, we suggest that, in order to hedge the risks in the international oiltrade, China should play an active role in the energy diplomacy strategy, build andkeep strategic relationship with oil sources, and put the diversification strategy intopractice.This paper contributes in3ways. First, we analyse the risks factors in theinternational oil trade in a systematic and quantitative measure. Second, after literacyreviewing and consulting the specialists, we build up a multi-layer herarchy system ofrisk factors. Third, we revise the consistency problem in traditional AHP method withgenetic algorithm, by integrating GA into AHP.Although this paper makes some progress in AHP method to analyze risks of oiltrade, it has some limitations. AHP method requires that the influence between factorsin different layers is one-way directional, that is, from bottom to up. The effectivenessof traditional AHP method or even GA-AHP method is doubtful in network structureand feedback relations. However, network structure and feedbacks between factorsexists in one way or another. Dealing with network structure and feedback relations isthe next objective of the author.更多还原