大数据环境下政府投资建设项目决策模型研究

发布时间：2018-01-02 00:14

  本文关键词：大数据环境下政府投资建设项目决策模型研究　出处：《华北电力大学(北京)》2017年博士论文　论文类型：学位论文

  更多相关文章： 政府投资建设项目 大数据环境 决策机制 概率区间数 随机占优度

【摘要】：政府投资建设项目是指为了实现政府职能,满足社会公众诉求,贯彻落实经济社会发展战略,利用国家财政预算内、外资金,或以财政性资金作为还款来源的借贷性资金建设的项目。大量上马的政府投资建设项目对于改善居民生活环境,促进经济社会发展,保证经济平稳增长等方面意义重大,但是同时也引起了社会各界对于政府投资建设项目决策科学性的高度关注和潜在担忧。大数据技术的飞速发展为我们解决传统问题提供了新的思路,越来越多的学者看好大数据技术在政府投资建设项目决策领域的应用潜力。然而,由于政府决策者对于大数据技术还比较陌生,无法运用大数据的思维思考决策问题,构建大数据环境下全新的决策逻辑框架,提出研发诉求;而熟悉大数据技术的研发人员又不了解现实的决策环境和决策人的诉求,无法提供对应的需求响应方案。在大数据技术人员和政府决策者间存在一条无法逾越的知识鸿沟,制约着大数据技术在政府投资决策领域内的应用。为了填补这条横在政府决策者与大数据技术人员间的知识鸿沟,本文首先对当前的大数据采集、储存、处理、决策支持技术的发展现状进行了系统的梳理,厘清了大数据技术的技术边界。随后在可以实现的大数据技术范围内,深入分析了大数据环境下政府投资建设项目的决策情境,并进一步提出了全新的大数据环境下的政府投资建设项目决策机制。最后提出了在全新的决策环境和机制下的决策模型。具体研究内容包括以下六个部分:(1)政府决策中的大数据技术应用研究。本文首先采用文献分析法,从大数据采集技术、大数据存储技术、大数据处理技术和大数据决策支持技术四个方面,对当前可用的大数据技术进行全面的梳理和分析,从而明确大数据技术的技术边界,为之后的研究奠定技术基础。(2)大数据环境下政府投资建设项目决策情境研究。在明确大数据技术边界的基础上,本文进一步研究了当前的大数据技术环境下,政府决策思维应该进行怎样的转变以及如何进行转变。随后通过构建公众参与有效决策模型,分析了大数据技术对公众参与决策带来的巨大影响。并讨论了政府投资建设项目决策中的数据种类、数据来源、数据分析方法以及数据安全问题。(3)大数据环境下政府投资建设项目决策机制设计。该部分研究基于大数据环境下政府投资建设项目决策的全新情境,以决策的科学化、民主化、合理化为目标,讨论并重新定义了政府、公众和专家在政府投资建设项目决策中的角色职责,并以此为基础设计大数据环境下政府投资建设项目决策机制。(4)基于近似随机占优的大数据决策模型研究。根据大数据环境下政府投资建设项目决策的数据形式的特征,本研究提出了可以同时分析和处理实数、随机数和区间数三类决策数据的算法理论——概率区间数及其运算规则。在此基础上基于近似随机占优理论,提出一种考虑了所有公众风险态度和效用偏好的决策模型,该模型能够高效率低成本地处理超大决策群体的风险态度和效用偏好,在保证决策精度的同时,大大降低对大数据处理技术的要求。(5)政府与公众异构偏好集结决策模型研究。针对政府和公众给出的异构偏好,本文通过设计模型将政府决策者加工过的偏好信息还原为较原始的状态,把公众与政府决策者的异构偏好转化为同一形式,实现不同类型决策者异构偏好的集结。此外,针对决策中有多个政府主体参与决策的情况,本文还进一步提出了多个决策主体的异构偏好集结模型。(6)政府与公众偏好趋同靶向调整模型研究。本文首先提出了基于Kendall和谐系数和基于修正前后的指标偏好权重向量间欧式距离的两个满意度评估模型。若认为满意度不能满足决策要求,则需要对决策者的偏好进行调整。针对大数据环境下政府投资建设项目决策的特点,本文构建了面向不同调整阶段的政府与公众风险和指标偏好调整模型,借助数学算法辅助决策双方根据自身意愿对偏好进行精准快速的调整。
[Abstract]:The construction project of government investment is that in order to achieve the functions of the government, to meet the public demands, implement the strategy of economic and social development, the state budget, funds, or financial capital as a source of repayment of loan funds for construction projects. The government launched a large number of investment projects to improve the living environment, promote the economic society the development, guarantee the steady economic growth of great significance, but also aroused great concern for government investment in the construction of the scientific decision making of the project and potential concerns. The rapid development of information technology provides a new way for us to solve the traditional problems, more and more scholars are optimistic about the potential application of big data technology in the field of government decision making the investment in construction projects. However, because of government decision makers for big data technology is still relatively unfamiliar, not the use of big data thinking Dimensional thinking decision problem, construct the decision logic framework of the new big data environment, put forward a new demand; and familiar with big data technology R & D and do not understand the real decision-making environments and people's demands, to provide the corresponding demand response program. There is an insurmountable gap in knowledge and technical personnel and government big data decision makers, restricting the application of big data technology in government investment decisions in the field. In order to fill the knowledge gap across government policymakers and big data technology personnel, based on the current data collection, storage, processing, sorting out the system development present situation of decision support technology, clarify the technology of boundary data technology. Then in the big data technology range can be achieved within the in-depth analysis of the data under the environment of government investment construction project decision-making situation, and further. The decision mechanism of government investment construction project under the new environment of big data. Finally put forward the decision model in the decision-making environment and new mechanism. The specific research contents include the following six parts: (1) research on big data technology application in government decision-making. Firstly, using the method of literature analysis, from big data acquisition technology, data storage technology, four aspects of large data processing technology and data decision support for comprehensive analysis and analysis of the currently available data technology, so as to clear the boundary technology of data technology, and lay a foundation for the study. (2) after the big data environment government research project making investment and construction. Based on big data technology boundary, this paper further studies the big data technology under the current environment, government decision-making should be how to change and how to Change. Then through the construction of public participation in the effective decision-making model, analysis of large data technology of public participation in the huge impact brought by decision-making. And discussed the construction of government investment project decision of data types, data sources, data analysis and data security issues. (3) big data under the environment of government investment construction project decision-making mechanism design. This part of the study based on the new situation of the government under the big data environment construction project investment decision, to the scientific decision-making, democratic, rational discussion as the goal, and a new definition of the government, the public and experts in the construction project of government investment decision-making in the role, and on the basis of the design of large data under the environment of government investment construction the project decision-making mechanism. (4) research on big data approximate decision model based on stochastic dominance. According to government data environment construction project investment decision according to the number of forms The characteristics, this study proposes can also analyze and deal with real numbers, random number and interval number decision-making data algorithm theory: the probability interval number and its operation rules. Based on this approximation based on stochastic dominance theory, proposes a decision-making model considering all public risk attitudes and preferences, the model is capable of high efficiency and low cost processing of large groups of decision risk attitudes and preferences, in which the decision accuracy at the same time, greatly reduce the large data processing requirements. (5) the government and the public on the heterogeneous preference decision model. According to the preference of the government and the public are heterogeneous, the design model of government decision makers processed preference information reduction as compared to the original state, the heterogeneous preference public and government decision makers into the same form, different types of decision makers of heterogeneous aggregation of preference. In addition, according to the decision of a number of government participation in decision making, this paper further puts forward the main decision-making multiple heterogeneous preference aggregation model. (6) to study the adjustment model of the government and the public preference is given in this paper. The convergence of the target two satisfaction index preference weight vector based on Euclidean distance based on Kendall coefficient and harmony before and after the correction between the satisfaction evaluation model. If that can not meet the demand of decision, requires policymakers preference adjustment. According to the characteristics of big data in government investment project decision-making environment, this paper constructs the government and the public and the risk index of preference adjustment model for different adjustment stage, with the help of mathematical decision algorithm according to both sides own preferences for accurate rapid adjustment.

【学位授予单位】：华北电力大学(北京)
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TP311.13;F282
，

本文编号：1366837