【作者】 张艺；

【导师】 余开朝；

【作者基本信息】 昆明理工大学 ， 工业工程， 2018， 硕士

【摘要】 制造业是一个国家或地区国民经济发展的重要支柱,其发展水平反映了一个国家或地区的综合实力。2008年全球金融危机爆发后,美国、德国、英国、日本等世界发达国家纷纷出台了以发展智能制造为主要特征的国家制造业发展战略,通过智能制造重塑制造业,重振实体经济。为实现从制造大国向制造强国转变,我国政府也于2015年5月提出了以智能制造为主攻方向的“中国制造2025”,大力推进智能制造。在我国智能制造的推进中,如何对我国制造业智能制造能力进行评价和快速有效的提升,是各级政府和企业所关心的一个问题。通过制造业智能制造的能力评价,有利于企业了解其智能制造发展水平,明确企业智能制造具备的基础条件、具有的优势和存在的不足,为企业制定智能制造发展规划和智能化技术改造决策提供依据,也为各级政府及行业主管部门提供一个指导、考评本地区制造业企业智能制造发展水平的工具。因此,智能制造能力评价是我国推进智能制造中一个不可避免的问题,值得深入开展研究。本文以我国23个主要省市的智能制造能力为研究对象,参考前人的研究成果,以国家统计局、工信部等官方发布的反映我国智能制造发展水平的相关指标和数据为基础,侧重于生产制造过程开展智能制造能力评价研究。首先,通过对因子分析法、层次分析法和人工神经网络等常用评价方法的分析比较,选择BP神经网络对我国主要省市的智能制造能力进行评价;其次,利用统计分析等技术从产品创新能力、信息化水平和产品流通能力三个方面筛选出申请专利数、平均企业物联网覆盖率和平均拥有等级公路里程等20个二级指标,通过因子分析构建了一种智能制造能力评价指标体系。通过对BP神经网络的不同学习算法进行训练,选择带自适应学习率和动量因子的梯度下降法建立BP神经网络模型,并确定最佳的隐含层神经元个数;最后,将20个评价指标的23组原始样本数据输入模型进行仿真。通过仿真得到本文所选23个评价对象的智能制造能力评价值,为本地区智能制造的发展水平提供了决策参考。更多还原

【Abstract】 Manufacturing industry is an important pillar of national economic development in a country or region.The level of manufacturing industry development reflects the comprehensive strength of a country or region.After the outbreak of the global financial crisis in 2008,developed countries such as the United States,Germany,the United Kingdom,and Japan have all introduced a national manufacturing development strategy that focuses on the development of intelligent manufacturing.Through intelligent manufacturing,it has reshaped the manufacturing industry and revitalized the real economy.In order to realize the transformation from the manufacturing country to the manufacturing power,the Chinese government also proposed “Made in China 2025”,which focuses on intelligent manufacturing in May 2015 to vigorously promote intelligent manufacturing.In the advance of China’s intelligent manufacturing,how to evaluate and improve the intelligent manufacturing capabilities of China’s manufacturing industry is a concern for governments and enterprises at all levels.Through the evaluation of intelligent manufacturing capabilities of manufacture,it is beneficial for enterprises to understand the level of intelligent manufacturing development,to clarify the basic conditions,advantages,and deficiencies of enterprise intelligent manufacturing,and to provide the basis for the enterprise to formulate intelligent manufacturing development planning and intelligent technical transformation decision.Based on this,it also provides tools for governments and industry authorities at all levels to provide guidance and evaluate the level of smart manufacturing development in manufacturing companies in the region.Therefore,the evaluation of intelligent manufacturing capability is an unavoidable issue in the promotion of smart manufacturing in China,which is worth further research.This thesis took the smart manufacturing capabilities of 23 major provinces and cities as the research object.Based on the research results of previous generations,it is based on relevant indicators and data released by the National Bureau of Statistics and the Ministry of Industry and Information Technology that reflect the level of China’s smart manufacturing development,focus on the manufacturing process to carry out intelligent manufacturing capacity evaluation research.Firstly,through the analysis and comparison of factor analysis,analytic hierarchy process and artificial neural network and other commonly used evaluation methods,BP neural network was selected to evaluate the intelligent manufacturing capabilities of major provinces and cities in China.Secondly,20 secondary indicators such as the number of applied patents,the average corporate Internet of Things coverage,and the average number of graded highway miles were selected through statistical analysis and other technologies from the aspects of product innovation capability,information level,and product circulation capability.An intelligent manufacturing capability evaluation index system has been constructed.By training different learning algorithms of BP neural network,a BP neural network model is established by gradient descent method with adaptive learning rate and momentum factor,and the number of optimal hidden layer neurons is determined.Finally,23 sets of original sample data of 20 evaluation indicators were input into the model for simulation.Through the simulation,the intelligent manufacturing capability evaluation values of the 23 evaluation objects selected in this paper are provided,which provides decision-making reference for the development level of intelligent manufacturing in the region.更多还原