控制图在检测实验室质量控制中的应用

0前言 在ISO/IEC 17025<检测和校准实验室能力的通用要求>中,技术要求的第9个要素"检测和校准结果质量的保证"(即5.9条)要求实验室在实际可行的情况下采用统计技术对检测结果数据进行技术核查,以监控检测的有效性,预防问题的发生.目前,各检测实验室正开始积极探索统计技术在检测质量控制方面的应用.本文通过运用 (-X)-R控制图分析钢勾心抗弯刚度的检测数据,说明控制图在检测实验室质量控制中的应用.     

具有可变抽样区间的残差MEWMA控制图

为了提高控制图监控多元自相关过程小波动的效率,研究具有可变抽样区间(VSI)的残差MEWMA控制图。首先,对残差MEWMA控制图进行可变抽样区间设计,用蒙特卡罗模拟方法计算其平均报警时间。其次,对所设计的可变抽样区间残差MEWMA与固定抽样区间的残差MEWMA进行比较研究。结果表明,可变抽样区间残差MEWMA控制图监控效率优于固定抽样区间的残差MEWMA控制图,能更快地发现过程中的变化。然后,通过对比研究得出,VSI残差MEWMA控制图的受控性能优于VSI MEWMA控制图。最后,研究平滑系数对过程监控性能的影响,得出过程波动较小时,采用较小的平滑系数;分析自相关系数对过程监控性能的影响关系,得出对于相同波动,相关系数越小,VSI MEWMA残差控制图的监控效果越好。     

控制图在实验室检测中的应用

本文将统计质量控制方法应用于实验室检测过程中,通过建立检测过程控制图,对检测过程进行评估、监控,判断检测过程是否存在异常,使检测结果得到有效控制。     

监控不合格品率的MEWMA-p控制图设计

传统Shewhart-p控制图只对单一属性的不合格品率进行监控,在过程发生偏移时有一定的滞后性。为提高不合格品率控制图的精度,提出一种多元指数加权移动平均不合格品率(multivariate exponentially weighted moving average p, MEWMA-p)控制图。该控制图将多个属性的不合格品率应用于多元指数加权移动平均控制图,可同时对多个属性进行监控,并且对于小范围的偏移更加敏感。对比分析同等偏移程度下指数加权移动平均不合格品率(exponentially weighted moving average p, EWMA-p)控制图与MEWMA-p控制图的平均运行长度(average run length,ARL)结果,并通过模拟仿真说明该方法的有效性。     

基于APL的偏态控制图优化设计

以平均产品长度(APL)为评价控制图性能的标准,研究了偏态控制图的优化设计问题.针对一般控制图无法有效解决偏态总体的不对称性的情况,采用赋权方差法来构造非对称的偏态控制图,并获得其最优设计模型;最后给出了模型的灵敏度分析及算例.     

基于多点报警准则的高质量过程缺陷率控制图

为了提高用于高质量过程监控的累积计点值控制图的监控效率,提出了一类采用多点报警准则的高质量过程缺陷率控制图设计,并给出了计算这类改进型控制图的平均运行长度的马尔科夫链方法。为了验证该设计方法的改进效果,分析比较了采用多点报警规则的3种累积计点值控制图（报警规则为连续两点出界、连续三点中有两点出界、连续三点出界）与报警规则不改变的（报警规则为有一个点出界）累积计点值控制图的监控效率。比较结果表明：在受控状态下平均运行长度都为370的情况下,多点报警的累积计点值控制图的失控状态下的平均运行长度分别较报警规则不变的控制图的平均运行长度减小了约44%、40%、63%,控制图发现缺陷率增大的效率明显改善。得出结论：这种多点报警的累积计点值控制图发现缺陷率增大的速度提高40%~60%。     

质量控制图在实验室质量管理中的应用

质量控制图是实验室开展质量控制工作的一种重要工具,便于实验室有效识别检测过程中出现的系统性变异,及时采取措施,使检测过程保持稳定。     

平稳自相关过程的EWMA控制图

讨论了对平稳自相关过程中出现的较小波动进行监控的一种方法.采用自回归移动平均(ARMA)模型对平稳自相关过程进行适当的拟合,通过计算残差的方法消除过程中的自相关要素,并在此基础上提出对于均值和方差出现的较小波动进行监控的指数加权移动平均(EWMA)控制图的构造.通过与其它几种方法的比较来说明该方法在监控平稳自相关过程时有更好的效率.     

基于谱残差法显著图的视频镜头边界检测

目的针对传统视频镜头边界检测算法精度低和较复杂等缺点,提出一种基于谱残差显著图和分块灰度直方图相结合计算似然比的算法。方法为了提取有效结构信息作为特征来比较帧间差异,采用谱残差方法得到各帧图片的显著图。将显著图分块并计算各块的灰度直方图数据,进一步提高检测的精度。计算各帧似然率后用文中提出的帧间差指标计算比较帧间差异。结果所提方法的综合检测精度能达到93.82%,对常见影响因素下相邻未变化帧的相似性保持在99%以上。结论实验结果表明,文中算法的检测精度高于传统方法,过程相对简单,对常见影响因素具有较强的鲁棒性。     

提高超声波检测系统信噪比的理论研究及检测控制方案

针对超声波检测系统的特点,从实际应用出发,对系统进行理论分析,得到了换能器的最佳驱动信号,大大提高了系统的信噪比,为超声波检测系统的实际应用提供了有用的理论依据;同时,还提出了系统检测控制可行性方案,很大程度上解决了超声波检测系统的噪声问题。实验结果表明,该理论正确,控制方案实用。     

Effects of Model Accuracy on Residual Control Charts

Residual control charts are acknowledged to be effective tools for statistical process control of multistage processes. In these monitoring procedures, the models on the stage‐wise correlation should be first derived before the control charts are implemented. Therefore, the monitoring performance is inevitably affected by the model fitting scheme. Most of the previous works are under the assumption that the derived models represent the process behavior perfectly. Far less is known about the effects of the model inaccuracy on the monitoring performance. To investigate the effects of the underlying models on the monitoring performance, residual control charts based on two different modeling schemes are compared in this paper. The results indicate that the charting performance is correlated with the model fitting schemes. That is, a more accurate model will significantly increase the detection power and decrease the false alarm rate as well. Copyright © 2015 John Wiley & Sons, Ltd.     

Conditional Control Charts for Weibull Quantiles Under Type II Censoring

In this article, we propose control charts for the quantiles of the Weibull distribution, for type II censored data, based on the distribution of a pivotal quantity conditioned on ancillary statistics. These control charts must be considered as alternatives to bootstrap type control charts. We derive an analytical form of the conditional distribution function of the monitored statistic and we use this function to propose ARL‐unbiased control limits. We further demonstrate that the proposed conditional chart have a general analytical form for the ARL that can be evaluated numerically without use of simulations and we also show that these charts perform at least as well as the bootstrap type ones. We finally apply the conditional charts to a dataset on the strength of carbon fibers to detect shifts in a specified Weibull quantile. Copyright © 2014 John Wiley & Sons, Ltd.     

Monitoring the Weibull Shape Parameter by Control Charts for the Sample Range of Type II Censored Data

In this paper, we propose control charts to monitor the Weibull shape parameter β under type II (failure) censoring. This chart scheme is based on the sample ranges of smallest extreme value distributions derived from Weibull processes. We suggest one‐sided (high‐side or low‐side) and two‐sided charts, which are unbiased with respect to the average run length (ARL). The control limits for all types of charts depend on the sample size, the number of failures c under type II censoring, the desired stable‐process ARL, and the stable‐process value of β. This article also considers sample size requirements for phase I in retrospective charts. We investigate the effect of c on the out‐of‐control ARL. We discuss a simple approach to choosing c by cost minimization. The proposed schemes are then applied to data on the breaking strengths of carbon fibers. Copyright © 2011 John Wiley & Sons, Ltd.     

On the Performance of Two‐sided Control Charts for Short Production Runs

In this paper, statistical performance measures for short‐run production are summarized and proposed. Evaluation and computation based on these performance measures were developed for the two‐sided Shewhart, cumulative sum, and exponentially weighted moving average control charts. Applying these measures, we provided numerical illustrations to compare the efficiency of the three charts and the corresponding Q charts in the cases of known and unknown nominal values of process parameters. Copyright © 2011 John Wiley & Sons, Ltd.     

Performance of Tukey's and Individual/Moving Range Control Charts

This paper compares two control charts: Tukey (TCC) and individual/moving range (XmR) control charts. Both are designed to examine single observation per time period, but little is known about which one is more efficient and under what conditions. We simulated data from different distributions and examined the performance of the two control charts on these data. Performance was assessed using the of average run length, extra quadratic loss, median run length, standard deviation run length, performance comparison index, and relative average run length. Overall, TCC was more efficient than XmR, when observations had binomial, Rayleigh, logistic, lognormal, Maxwell, normal, Poisson, Weibull (with α = 10, β = 1), and Student's t (30 and 10 degrees of freedom) distributions. XmR was more efficient when observations had Student's t (with 4 degrees of freedom) and gamma (with α = 4, β = 1) distributions. These results suggest that improvement teams could reach faster conclusions if they use TCC in most common situations. Copyright © 2014 John Wiley & Sons, Ltd.     

Bayes‐conditional Control Charts for Weibull Quantiles under Type II Censoring

In this paper, we develop a Bayesian approach for monitoring Weibull quantiles under Type II censoring when prior information is negligible relative to the data. The posterior median of quantiles is considered as the monitored statistic. A method based on the relationship between Bayesian and conditional limits under an appropriate prior distribution is proposed to obtain the posterior median of quantiles in closed form. A pivotal quantity based on the monitored statistic is proposed, and its distribution is conditionally derived. Then, the Bayes‐conditional control limits are proposed. For the proposed charts, the probability of out‐of‐control can be derived without use of simulation. The performance of the Bayes‐conditional charts is compared with the bootstrap charts through the simulation methods. The results show that to monitor the first quantiles, the lower‐sided Bayes‐conditional charts perform better than bootstrap charts in detecting a downward shift caused by decreasing in the shape parameter. Finally, an illustrative example is provided. Copyright © 2016 John Wiley & Sons, Ltd.     

Control Charts for the Lognormal Mean

Among a set of tools that form the core of statistical process control, statistical control charts are most commonly used for controlling, monitoring, and improving processes. The conventional control charts are based on the assumption that the distribution of the quality characteristic to be monitored follows the normal distribution. However, in real applications, many process distributions may follow a positively skewed distribution such as the lognormal distribution. In this study, we discuss the construction of several control charts for monitoring the mean of the lognormal distribution. A real example is used to demonstrate how these charts can be applied in practice. Copyright © 2015 John Wiley & Sons, Ltd.     

MEWMA Control Chart and Process Capability Indices for Simple Linear Profiles with Within‐profile Autocorrelation

A multivariate exponentially weighted moving average (MEWMA) control chart is proposed for detecting process shifts during the phase II monitoring of simple linear profiles (SLPs) in the presence of within‐profile autocorrelation. The proposed control chart is called MEWMA‐SLP. Furthermore, two process capability indices are proposed for evaluating the capability of in‐control SLP processes, and their utilization is demonstrated through examples. Intensive simulations reveal that the MEWMA‐SLP chart is more sensitive than existing control charts in detecting profile shifts. Copyright © 2016 John Wiley & Sons, Ltd.     

On the Performance of Control Charts for Simultaneous Monitoring of Location and Dispersion Parameters

A single control chart is very famous to control assignable causes that shift the process because of variations in parameters (e.g., location and dispersion). Simultaneous monitoring of processes is another popular approach used for the bilateral processes. In this study, we have proposed the mixed control charts for simultaneously monitoring of process location and dispersion parameters. We have used the idea of mixed exponential weighted moving average and cumulative sum charts and designed the charting structures for simultaneous monitoring. The proposals are compared with several existing counterparts. The comparisons reveal numerous advantages of the proposed charts over the other existing scheme. The practical application of the proposed charts is also highlighted using an illustrative example based on a real dataset. Copyright © 2016 John Wiley & Sons, Ltd.     

Improved Fast Initial Response Features for Exponentially Weighted Moving Average and Cumulative Sum Control Charts

Exponentially weighted moving average (EWMA) and cumulative sum (CUSUM) control charts have found extensive applications in industry. The sensitivity of these quality control schemes can be increased by using fast initial response (FIR) features. In this paper, we introduce some improved FIR features for EWMA and CUSUM control charts and evaluate their performance in terms of average run length. We compare the proposed FIR‐based EWMA and CUSUM control schemes with some existing control schemes, that is, EWMA, FIR‐EWMA, CUSUM, and FIR‐CUSUM. It is noteworthy that the proposed control schemes are uniformly better than the other schemes considered here. An illustrative example is also given to demonstrate the implementation of the proposed control schemes. Copyright © 2013 John Wiley & Sons, Ltd.