PCI理论在QS公司焊机生产中的应用研究

发布时间：2018-05-12 17:02

本文选题：质量控制 + 过程能力指数　；参考：《天津科技大学》2017年硕士论文

【摘要】：当下,随着实体经济的繁荣发展,客户对于产品的要求日趋严格,如何在激烈的市场竞争中站稳脚跟一直是企业所思考的问题。本文围绕QS公司焊机生产的质量控制方面进行深入研究。在对生产进行质量控制时,计算过程能力指数(PCI)进而分析生产系统的生产过程能力十分重要。传统情况下都视过程输出数据服从正态分布这一理想状态,且进行过程能力分析时,不考虑生产模式,皆采用相同的分析方法。针对以上问题,本文以统计过程控制理论为基础,依托QS公司焊机生产,在计算过程能力指数时,充分考虑数据非正态性与QS公司多品种小批量生产模式,建立多品种小批量模式下非正态过程能力指数的计算方法,并应用于QS公司焊机生产以验证其有效性和可行性。首先,分析非正态过程能力指数的优劣。通过充分了解过程能力相关理论的基础上,建立常用的五类非正态过程能力指数计算的数学模型。利用蒙特卡洛方法,选取对数正态分布、威布尔分布数据带入各数学模型,并对各参数进行赋值。借助Matlab仿真输出模拟结果Cpu,对比分析得到结果:在实际的应用过程中由于会受到诸多因素的限制,故采用偏态过程能力指数Cs方法更为理想。其次,建立多品种小批量生产模式下的过程能力指数计算方法。针对传统统计过程控制的局限性,提出Bootstrap相关理论与计算方法,同时重点研究了基于Bootstrap相关方法的过程能力指数及其置信区间的计算。建立数学模型并编写Matlab程序。通过上下控制限、样本容量、抽取Bootstrap随机替换样本的样本容量、抽取个数等参数的设定,进行仿真,并将得到的估计值、置信区间以及区间宽度进行对比,得到结论:t百分位数Bootstrap(Percentile-t Bootstrap,PTB)方法较另外两种方法在处理多品种小批量问题上更具优势,得到的过程能力指数估计值与置信区间更加精确。最后,进行QS公司焊机生产的过程能力指数分析。选取DNT3-160通体悬挂点焊机的C型点焊钳为研究对象,建立QFD质量屋确定关键质量特性。通过正态性检验、结合偏态过程能力指数Cs与t百分位数Bootstrap方法,得到质量特性值的过程能力指数。最终提出针对于QS公司过程能力分析的一般流程,依据此流程改善QS公司过程能力分析的手段,实现该企业焊机生产的质量控制水平。
[Abstract]:At present, with the prosperity of the real economy, the requirements of customers for products become increasingly strict, how to stand firm in the fierce market competition has been considered by enterprises. This paper focuses on the quality control of QS welder production. In the process of quality control, it is very important to calculate the process capability index (PCI) and then analyze the production process capability of the production system. Under the traditional condition, the process output data is normally distributed from the ideal state, and the same analysis method is adopted in the process capability analysis, regardless of the production mode. In order to solve the above problems, based on the statistical process control theory, this paper relies on the production of welding machines in QS Company. When calculating the process capability index, we fully consider the non-normality of data and the multi-variety and small-batch production mode of QS Company. The calculation method of non-normal process capability index in multi-variety and small-batch mode was established and applied to the production of QS welder to verify its validity and feasibility. Firstly, the advantages and disadvantages of the capability index of non-normal process are analyzed. On the basis of fully understanding the relevant theory of process capability, the mathematical models of five kinds of non-normal process capability exponents are established. Using the Monte Carlo method, the logarithmic normal distribution is selected, and the Weibull distribution data are brought into each mathematical model, and the parameters are assigned. By using Matlab simulation to output the simulation results, the results are compared and analyzed. The results are as follows: in the practical application process, due to the limitation of many factors, it is more ideal to use the skewness process capability index Cs method. Secondly, the calculation method of process capability index in multi-variety and small-batch production mode is established. In view of the limitation of traditional statistical process control, Bootstrap correlation theory and calculation method are put forward, and the calculation of process capability index and its confidence interval based on Bootstrap correlation method is studied emphatically. Build mathematical model and write Matlab program. By setting up and down control limit, sample size, sample size of Bootstrap random replacement sample, extracting number and so on, the simulation is carried out, and the estimated value, confidence interval and interval width are compared. It is concluded that the ratio t percentile Bootstrap(Percentile-t Bootstrapper PTB method is superior to the other two methods in dealing with multi-variety and small-batch problems, and the estimation of process capability index and confidence interval are more accurate. Finally, the process capability index analysis of QS welder production is carried out. The C type spot welding clamp of DNT3-160 universal suspension spot welder was selected as the research object, and the QFD quality house was established to determine the key quality characteristics. The process capability index of mass characteristic value is obtained by normal test combined with skew process capability index Cs and t percentile Bootstrap method. Finally, the general process of process capability analysis in QS company is put forward. According to the process capability analysis method, the quality control level of welding machine in QS company is realized.
【学位授予单位】：天津科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F224;F426.4;F273

【参考文献】