机械制造企业的数据挖掘应用研究

VIP免费
3.0 牛悦 2024-11-19 4 4 867.11KB 62 页 15积分
侵权投诉
摘 要
目前数据挖掘技术在我国机械制造业中的应用还处于起步阶段，而现有的研
究大多数仅仅局限于企业 ERP 管理系统的信息化建设，而忽视了该信息系统所产
生的大规模业务数据及其所蕴含的大量潜在价值信息。该类信息对优化业务管理，
高效分析市场需求走向，提高企业竞争力将会起到越来越重要的作用。因此，将
数据挖掘技术充分而高效的运用于企业数据的分析和管理，是现阶段我国机械制
造业信息化发展的关键技术领域。
制造业的产品质量管理系统是企业发展的基石，也是提升公司品牌效应的重
要因素。结合数据挖掘技术，对质量管理决策效率进行优化，规范管理制度，对
企业发展起着至关重要的作用。
本文研究的关键问题主要包括：对机械制造业产品质量管理业务中存在的严
重问题进行深入分析；运用广义知识挖掘方法监督业务流程，掌握问题发生情况；
通过改进，使 Apriori 算法由单维属性研究向多维属性研究转变，并将其应用于多
维属性间的关联知识挖掘；改进 ID3 算法，构建决策支持模型，以分类预测挖掘
方法对业务流程的效率进行优化等。
首先，深入机械制造企业，结合该业务的 ERP 管理系统，对产品质量管理业
务进行调研。获取业务属性、业务流程、数据流图、各部门职责等相关信息，并
结合调研情况详细分析了该业务管理中所存在的问题。
其次，根据业务需求，结合数据挖掘广义知识模式，对相关数据进行了分析
概括，并以可视化技术展现分析结果，以帮助管理者监督业务流程，掌握问题发
生情况。
然后，运用基于 Apriori 算法的多维关联规则挖掘方法，对业务属性之间的关
联性展开研究，并生成了强关联规则。经评价分析，该类规则能够引导相关部门
进行责任归属问题的分析，优化质量管理，并对某些问题的发生起到预防作用。
最后，在认真分析导致业务堆积、流转效率偏低原因的基础上，构建了一种
基于 ID3 算法的决策支持模型。该模型以标准的 CRISP-DM 过程进行构建，并利
用基于信息增益率的计算分类方法，对 ID3 算法进行了改进。通过测试分析，该
模型具有较高准确率，能有效提高业务流转效率。
关键词： 机械制造业 数据挖掘 CRISP-DM 决策树 关联规则
ABSTRACT
At present, the application of data mining technology in China's machinery
manufacturing industry is still in its infancy phase.Most of the currently research just
focus on the ERP management system’s information construction,while ignoring the
large-scale business data which produced by the information system, and the data
contains large number of potential valuable information.Such information will play a
more and more important role in optimize business’s management, analysis market’s
requirement and improve enterprises’ competitive powers.Therefore, make the data
mining technology more fully applied to the enterprise’s management and analysis, is a
key point of technology research in China’s machinery manufacturing.
The product quality management system is the cornerstone of the manufacturing
industry, and is also an important factor to enhance the brand of a company. Optimize
the quality management’s efficiency and standardize management system based on
data mining technology plays a critical role in enterprises’ development.
In this paper, the key issues include: take a deep analyze of the main problems in
machinery manufacturing’s product quality management business; using the
generalized knowledge mining method to study the monitor of the business process
and the analysis of the problems’ happen conditions; improve the Apriori algorithm
and make it apllied to the multi-dimensional attributes’ knowledge mining; using the
classification prediction method to optimize the business process, improved the ID3
algorithm of decision tree, and build a decision support model.
First of all, do research to the machinery manufacturing’s product management
system based on company’s ERP management system. And obtain the business
properties, processes, and responsibilities of different departments, then next give an
analysis of the exist problems based on the information.
Secondly, according to the business’s requirements, summarize the relevant data
and show the results in visual table and diagrams.In order to help managers to control
the business’s circulation and problem’s occurrence.
Then, study the multidimensional association rule mining method based on Apriori
algorithm and give a research on the correlation of the business properties, and
generated several association rules. After analysis and evaluation, the rules can help to
analyze the responsibility attribution and make prevention of certain issues’ occuration.
Finally, in order to improve the quality management business’s efficiency, this
paper builds a decision-making support model based on ID3 algorithm. The model was
built based on CRISP-DM standard process, and improved the ID3 algorithem by
using the gain ratio calculation method. By test and ananysis, the model has a high
accuracy and can standardize the business process and speeding the decision-making
step efficiently.
Keywords: machinery industry, data mining, CRISP-DM, decision
tree, association rules
目录
摘 要
ABSTRACT
第一章 绪论 .....................................................................................................................1
§1.1 研究背景 ............................................................................................................1
§1.1.1 机械制造业的发展 .................................................................................1
§1.1.2 数据挖掘技术的兴起 .............................................................................2
§1.2 主要工作 ............................................................................................................3
§1.3 内容结构 ............................................................................................................4
第二章 数据挖掘技术综述 .............................................................................................5
§2.1 数据挖掘概述 ....................................................................................................5
§2.1.1 定义 .........................................................................................................5
§2.1.2 特点 .........................................................................................................6
§2.1.3 应用 .........................................................................................................7
§2.1.4 局限性 .....................................................................................................8
§2.2 数据挖掘的方法分类 ........................................................................................9
§2.2.1 广义知识型 ...........................................................................................10
§2.2.2 关联知识型 ...........................................................................................11
§2.2.3 分类预测型 ...........................................................................................13
§2.2.4 方法的评价 ...........................................................................................15
§2.3 数据挖掘的标准过程 .....................................................................................16
§2.3.1 一般过程 ...............................................................................................16
§2.3.2 业界标准 ...............................................................................................16
§2.3.3 CRISP-DM 过程 ....................................................................................17
第三章 机械制造业产品质量管理业务理解 ...............................................................20
§3.1 背景介绍 ..........................................................................................................20
§3.2 流程理解 ..........................................................................................................20
§3.2.1 业务流程 ................................................................................................20
§3.2.2 部门职责 ................................................................................................22
§3.2.3 数据流图 ................................................................................................22
§3.3 需求分析 ..........................................................................................................25
§3.3.1 数据字典 ...............................................................................................25
§3.3.2 存在问题 ...............................................................................................28
§3.3.3 解决方式 ...............................................................................................28
§3.4 广义知识挖掘 .................................................................................................29
§3.4.1 业务流程控制 ........................................................................................29
§3.4.2 问题类型分析 ........................................................................................30
§3.4.3 缺陷等级分析 ........................................................................................31
§3.4.4 责任部门分析 ........................................................................................31
§3.4.5 问题趋势分析 ........................................................................................32
第四章 基于 Apriori 算法的多维关联规则分析 .........................................................33
§4.1 引言 .................................................................................................................33
§4.2 多维关联规则算法 .........................................................................................33
§4.2.1 关联规则 ...............................................................................................33
§4.2.2 Apriori 算法及其改进 ........................................................................... 35
§4.2.3 多维关联规则 .......................................................................................36
§4.2.4 关联规则挖掘过程 ...............................................................................37
§4.3 业务理解 .........................................................................................................38
§4.4 关联规则的挖掘 .............................................................................................38
§4.4.1 数据预处理 ...........................................................................................38
§4.4.2 挖掘频繁谓词集 ...................................................................................40
§4.4.3 生成关联规则 .......................................................................................41
§4.5 关联规则的评价 .............................................................................................42
§4.6 小结 .................................................................................................................43
第五章 一种基于 ID3 算法的决策支持模型 .............................................................. 44
§5.1 引言 .................................................................................................................44
§5.2 决策树算法 .....................................................................................................44
§5.2.1 基本思想 ...............................................................................................44
§5.2.2 决策树构建过程 ...................................................................................45
§5.2.3 ID3 算法及其改进 ................................................................................ 45
§5.3 业务理解 ..........................................................................................................46
§5.4 决策支持模型的构建 .....................................................................................47
§5.4.1 数据预处理 ...........................................................................................47
§5.4.2 决策树生成 ...........................................................................................48
§5.4.3 模型准确率评估 ...................................................................................49
§5.5 基本部署与实现 .............................................................................................50
§5.6 小结 .................................................................................................................51
第六章 结束语 ...............................................................................................................52
§6.1 论文总结 .........................................................................................................52
§6.2 工作展望 .........................................................................................................53
参考文献 .........................................................................................................................54
在读期间公开发表的论文和承担科研项目及取得成果 .............................................57
致 谢 .............................................................................................................................58
摘要：

摘要目前数据挖掘技术在我国机械制造业中的应用还处于起步阶段，而现有的研究大多数仅仅局限于企业ERP管理系统的信息化建设，而忽视了该信息系统所产生的大规模业务数据及其所蕴含的大量潜在价值信息。该类信息对优化业务管理，高效分析市场需求走向，提高企业竞争力将会起到越来越重要的作用。因此，将数据挖掘技术充分而高效的运用于企业数据的分析和管理，是现阶段我国机械制造业信息化发展的关键技术领域。制造业的产品质量管理系统是企业发展的基石，也是提升公司品牌效应的重要因素。结合数据挖掘技术，对质量管理决策效率进行优化，规范管理制度，对企业发展起着至关重要的作用。本文研究的关键问题主要包括：对机械制造业产品质量管理...

展开>> 收起<<
机械制造企业的数据挖掘应用研究.pdf

共62页,预览7页

还剩页未读， 继续阅读

相关推荐

更多
立即下载
作者：牛悦 分类：高等教育资料 价格：15积分 属性：62 页 大小：867.11KB 格式：PDF 时间：2024-11-19

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务

作者详情

相关内容

更多

推荐作者

热门标签

/ 62
评分 收藏
分享
立即下载 VIP免费下载
客服
关注