返回
类型 基础研究 预答辩日期 2017-11-30
开始(开题)日期 2015-05-28 论文结束日期 2017-06-16
地点 东南院305 论文选题来源 国家社科规划、基金项目     论文字数 9.5 (万字)
题目 基于链路预测的我国产业网络结构优化研究
主题词 产业网络,链路预测,产业共生,系统动力学,结构优化
摘要 目前,我国经济增速放缓,从高速增长进入中高速增长新常态,并以稳增长、调结构、转方式和防风险为发展目标。其中,如何调整产业结构,促进产业持续健康发展,已成为我国面临的关键经济问题。众多产业间物质、能量与信息的循环流动,共同形成了复杂的产业网络。由于网络结构决定了经济系统功能与产业结构转型升级的方向,因此,在新一轮的经济发展形势下,研究我国产业网络结构优化问题,为形成高效、协调、可持续的产业结构优化发展提供了明确的方向和任务,具有十分重要的理论意义和实践价值。 本文综合运用复杂网络理论、地理信息系统、系统动力学模型以及应用统计学理论等工具,以我国产业生态经济系统为研究目标,研究基于链路预测的我国产业网络结构优化问题。由于产业网络结构非常复杂,为了厘清其中关系,遵循研究思路从简单到复杂的原则,本文首先从横向静态视角分析整体(我国)与部分(主体功能区)之间的产业网络结构,为随后研究基于链路预测的我国产业网络结构演化路径做好铺垫。然后从纵向动态视角研究我国产业网络结构变化特征,并从探究我国产业网络演化历程中,分析网络结构的优化路径。再次,本文将采用基于外部冲击下时序链路预测模型进行仿真研究,构建网络结构优化指数对仿真出来的结果进行评估,从而得到优化的我国产业网络结构,并推演出网络结构优化路径。基于其中的优化路径,文章最后以互联网与传统产业为对象,进行案例分析,并以系统动力学为研究工具,揭示存在于二者链路上各个变量之间的因果关系,仿真分析二者共生优化的合理政策冲击配置,提出具体的优化策略。本文研究的主要内容与结论如下: 首先,分析了我国产业网络结构特征,考虑到区域内资源承载力在各个子系统之间的差异,因此着力于我国产业生态经济系统和主体功能区进行了研究。得到:由我国和主体功能区两个网络的接近中心性、介数中心性以及结构洞三个变量的测度可知,化学工业、邮政、批发零售和金融业等在内的服务业在这两个网络中的作用非常重要;我国产业网络以及主体功能区的产业网络呈现出稀疏化的发展态势,各个网络内产业间的投入产出关系都较弱;“京津冀”优化开发区产业具有更加重要的中心位置,且比“哈长城市群”重点开发区的产业自主能力更强;主体功能区间存在单桥联系,优化开发区的重工业对重点开发区装备工业具有支撑作用;在重点开发区中,存在着由高耗能产业、建筑业和运输业组成的“强供给”中心区块,它是区域经济的推进器,将产业发展的动能传递给其余区块。这些研究将为下一章从动态视角分析产业网络的演化问题提供了静态比较,同时使其优化研究“有的放矢”。 其次,从动态的视角构建新的链路预测模型,分析我国产业网络结构的演化特征,识别其结构演化的路径和方向。得到:批发和零售业、金融业的度数中心性逐年提升;同时,“十二五”规划结束后,第三产业的作用逐渐增强,产业结构优化调整的效应从预测得到的产业网络中得到了充分体现;“凝聚子群”的分析结果表明:在“十一五”时期,制造型服务业逐渐形成并不断发展;“十二五”末期,批发和零售业不仅与第二产业紧密相连,而且与第一产业亦形成了最直接、紧密的联系;另外,在“十一五”时期,金融业与房地产业的经济联系最紧密;“十二五”期间,金融业又形成了与教育、文化、体育和娱乐业以及公共管理和社会组织经济联系最直接、紧密的联系。该章内容将为下一章进一步研究在外部冲击下的产业网络结构优化问题提供了分析工具。 第三,基于上一章构建的链路预测模型,针对我国产业网络结构优化问题,研究了进一步考虑在外部因素冲击作用下的时序链路预测模型,然后通过调节外部冲击力度,以及测算产业网络结构优化指数,得到优化的我国产业网络结构,再通过研究该网络结构中核心—边缘产业以及对产业关键路径的分析,得到:最优网络结构具有明显的核心-边缘特征,核心-边缘产业按产业附加值由高到低依次递减;产业关键路径上关联程度最大的产业从住宿和餐饮业转变为通用、专用设备制造业,再转变为信息传输、计算机服务和软件业,说明信息传输、计算机服务和软件业正成为拉动我国经济增长、带动产业转型升级、实现创业创新的“强劲引擎”。 最后,产业共生是实现产业网络结构优化的有效方式。本文以系统动力学为研究工具,构建了共生网络优化的系统动力学模型,对传统产业与互联网产业共生优化进行了案例分析。得到:传统产业与互联网产业的共生优化发展,在合理的政策冲击配置下,具有融合、重构和创新的特征;关键政策的合理配置,即战略性新兴产业投资政策、研发补贴政策、资源税政策和环境污染治理投资政策达到绿色情景时,可以实现我国传统产业与互联网产业共生优化发展的目标。 本文研究的创新之处主要体现在以下几个方面: (1) 构建了基于链路动态变化的产业网络预测模型,该模型不仅考虑了两个节点之间的相似性程度与共同邻居节点的影响,而且体现了前期网络和当前网络中链路权重的变化程度,以及链路权重变化的方向,克服了现有相似性链路预测模型存在的不足。 (2) 构建了基于外部冲击下时序链路预测模型,量化了政府干预政策等外部因素对产业网络结构结构优化的影响,得到了网络核心—边缘产业以及产业关键链路的优化路径,改进了现有研究仅从网络结构特征等宏观层面进行优化分析的局限性。 (3) 在明确传统产业和互联网产业之间共生优化发展的内涵和特征的基础上,剖析了两类产业共生优化发展的关键变量以及彼此之间的因果关系,构建了相应的系统动力学模型,设置了基准、黄色、蓝色和绿色四种目标情景,通过模型对政策冲击配置方案进行了仿真分析,最终得到了两类产业共生优化的最优政策冲击配置策略,是应用研究的创新。
英文题目 Optimization research on structure of China’s industrial network based on the link prediction
英文主题词 Industrial network; Link prediction; Industrial symbiosis; System dynamics; Structure Optimization
英文摘要 At present, China’s economic growth is slowing from high-speed growth to medium-to-high speed of growth, and steady growth, structural adjustment, transfer mode and risk prevention are the development goals. Among them, how to adjust the industrial structure and promote the sustained and healthy development of the industry has become a key economic problem faced by our country. The circulation of material, energy and information among many industries has formed a complex industrial network. The network structure determines the function of the economic system and the direction of transformation and upgrading of the industrial structure. Therefore, in the new round of economic development, the study of China’s regional industrial network structure optimization problem provides a clear direction and task for the formation of efficient, coordinated and sustainable industrial rational distribution,and it has a very important theoretical and practical significance. This paper uses the complex network theory, geographic information system, system dynamics and applied statistics to study optimization of regional industrial network structure based on link prediction. The research area is China’s industrial ecosystem. Because the regional industry network structure is very complex, in order to sort out the relationship, we should follow the research principle from simple to complex. This paper first compares and analyzes the regional industrial network structure heterogeneity between the whole (China) and the part (main function area) from the perspective of transverse static view. It paves the way for further research on the evolution path of regional industry network structure based on link prediction. Second, this paper studies the change characteristics of regional industrial network structure from vertical dynamic angle of view, and analyzes the optimization path of network structure from exploring the evolution course of the regional industrial network. Third, simulation research is carried out in this paper based on external impact timing link prediction model. The regional network structure optimization index is used to evaluate the simulation results. Then we get the optimized regional industry network structure and push the network structure optimization path. Finally, the internet and traditional industries are taken as examples and the paper studies the causal relationship between variables in link, and simulates the policy matching relation, then putting forward optimization strategy. The main contents are as follows: Firstly, this paper analyzes the characteristics of the regional industrial network structure. Taking into account the differences in the carrying capacity of regional resources in each subsystem, the paper focuses on the comparative study of the industrial eco economic system and the main functional areas. The chemical industry, the service sector including the postal, wholesale, retail and financial sectors, is very important in these two networks. The two networks differ in three variables with closeness centrality, betweenness centrality, and structural holes. The industry density of the whole country is higher than that of the main functional area, but the relationship between the two inputs and outputs is not strong. The industry of optimizing development district of “Beijing-Tianjin-Hebei” has a more important center position and industrial autonomy than the industry of key-developing district of the “Harbin long city group”. There is a single bridge link between the main functional areas, and the heavy industry of the optimizing development district has a supporting role to equipment industry of key-developing district. There is a high-energy industry included construction and transportation industry which compose “strong supply” central block in the key-developing district. It is the regional economic propeller and transfers the industry development momentum to the rest of the block. These studies provide a static comparison of the evolution and optimization of industrial networks from a dynamic perspective. At the same time, the optimization on the network has a definite direction. Secondly, from the dynamic perspective, this paper propose a new link prediction model, analyze the evolution characteristics of the regional industrial network structure and identify the path and direction of its structural optimization. Taking the degree centrality in the index of industrial network as an example, the conclusions are: degree centrality which contains wholesale, retail trade and finance has been increasing year by year. The two industries belong to the tertiary industry. After the end of the 12th Five-Year Plan, the degree centrality of the tertiary industry is gradually increasing. The effect of the industrial structure optimization adjustment was fully reflected in the forecasted industrial network. The results of the analysis show that during the 11th Five-Year Plan period, the manufacturing service industry gradually formed and developed continuously. At the end of the 12th Five-Year Plan period, the wholesale and retail trade was not only close to the second industry connected, but also formed the most direct and strong relationship with the primary industry. In addition, in the “Eleventh Five-Year” economic development period, the financial sector is most closely linked to the real estate sector. During the “12th Five-Year” period, the finance has formed the most direct and strong relationship with education, culture, sports, entertainment industry, public management and social organization. This chapter will provide an analytical tool for further research on the optimization of industrial network structure under external shocks. Thirdly, according to a problem of the regional industrial network structure optimization, this paper constructs a time-series link prediction model with external impact. Then industrial network structure is optimized by adjusting the external impact strength and calculating index. The optimization direction of the network structure is obtained through studying the core-periphery industry and industrial critical path in the network structure. Finally, industrial symbiosis is an effective measure to realize the optimization of regional industrial network structure. A system dynamics model of regional industrial network symbiosis is constructed based on system dynamics. Based on the analysis of symbiosis optimization between traditional industry and Internet industry, this model considers the activities of industries in the regional industrial networks as a system. Based on the causal relationship between the variables in the regional industrial networks, the model simulates the contribution of industrial symbiosis to the optimization of regional industrial network structure. Through the analysis of policy scenarios, under different scenarios, the paper forecasts the contribution of industrial symbiosis to the regional industrial network GDP, the green decoupling index and ecological contribution to the development of efficiency. In addition, it discerns the main factors that influence the industrial symbiosis in the regional industrial network structure and puts forward the optimization Strategy. The innovations of this study are as follows: (1) This paper propose a prediction model of regional industrial network based on link dynamic changing when analyzing the regional industrial network structure and optimizing the path. The new link prediction model considers not only the similarity between the two nodes affected by a common neighbor, and considers but also the varying degree of link weight in the prophase and current network. On the basis of considering the degree of change, this paper further consider the direction of link weight changing, and overcome the shortcomings of similar link prediction model using only the current network to predict. (2) The paper build a time-series link prediction model with external impact and obtain optimized industry network structure. Then, the strategy of how to optimize the network structure is finally obtained by studying the core-periphery industry and the optimal path of the industry critical link. This paper improves the limitations of optimization analysis from the macro level of network structure in the existing research. (3) On the basis of the connotation and characteristics of the symbiotic and optimized development of traditional industries and internet industries, the paper analyzes the key variables and causal link between the two industries. The paper construct the corresponding system dynamics model and set the benchmark, yellow, blue and green four target scenarios. Through the simulation analysis of the allocation of policy, the paper finally obtained the optimal policy allocation of the two types of industrial symbiosis. The chapter is the innovation of applied research.
学术讨论
主办单位时间地点报告人报告主题
经济与管理学院 2014年6月12日 经济与管理楼B座4楼会议室 王斌 产业共生网络弹性、稳定性、脆弱性、柔性和适应性的对比分析与相关模型简介
经济与管理学院 2014年4月26日 九龙湖经济与管理学院A-208会议室 刘国强 高水平论文的写作与发表
南京师范大学经济与管理学院 2015年4月15日 南京师范大学学敏楼218 余菜花 环境规制会对产业布局产生显著的不利影响吗?
经济与管理学院 2013年 4 月20日 经济与管理楼B座4楼会议室 王斌 区域产业生态经济系统共生优化研究综述
经济与管理学院 2013年11月6日 经济与管理楼B座4楼会议室 王斌 中国产业生态系统研究评述
经济与管理学院 2015年10月25日 经济与管理楼B座4楼会议室 王斌 Study on Spatial Correlation Mechanism of Industries between Different Major Functional Areas Based on Grey Target Theory
经济与管理学院 2015年9月17日 经济与管理楼B座4楼会议室 王斌 产业生态系统中生态化和经济化的相互作用研究综述
经济与管理学院 2015年4月17日 经济与管理学院报告厅 贾良定 努力做一流的管理学研究
南京师范大学经济与管理学院 2016年3月17日 南京师范大学学敏楼218 张慧明 中国制造业如何走出“低端锁定”?
     
学术会议
会议名称时间地点本人报告本人报告题目
第28届中国控制与决策会议 2016年5月28日 中国.银川 Optimal Manufacturing Decisions Considering Capital Constraint
The 16th International Conference on Control, Automation and Systems 2016年10月17日 韩国.庆州 The Evolution Mechanism Analysis of Regional Industrial Network
     
代表作
论文名称
Research on Chinese energy consumption network between industrial sectors
Study on spatial correlation mechanism of industries between different major functional areas based
排污控制政策下企业的初始排污权采购策略
区域产业排污竞争网络的实证研究—以江苏省国有控股企业为例
 
答辩委员会组成信息
姓名职称导师类别工作单位是否主席备注
刘思峰 正高 教授 博导 南京航空航天大学
巩在武 正高 教授 博导 南京信息工程大学
袁健红 正高 教授 博导 东南大学
李爱国 正高 教授 博导 东南大学
陈伟达 正高 教授 博导 东南大学
      
答辩秘书信息
姓名职称工作单位备注
李绍芳 副高 副教授 东南大学