A note on economic sampling inspection plans

The determination of economic sampling inspection plans is discussed. A computational procedure that allows beta prior information is outlined.     

Economic in-line inspection sampling strategy with learning effects

This paper presents an innovative approach to a wafer inspection strategy that incorporates learning dynamics in semiconductor manufacturing factories. Using the data from fabrication lines (fabs) we demonstrate algorithms for computing the sampling strategy in terms of the percentage of wafers to sample for a process in different phases of a product life cycle. The average selling price and wafer starts per weeks are considered in the model. The paper provides an optimal solution methodology and concludes that the learning benefits of quality control activities may achieve the most cost-effective operations.     

Optimum rectifying inspection plans

The rectifying inspection plan is a useful tool in statistical quality control (SQC) to assure the average product quality between several stages of a production line. Two optimum rectifying inspection plans are presented. They minimize either ATI (average total inspection) or AOQ (average outgoing quality) and use the other as a constraint. The definitions for both ATI and AOQ are generalized to handle the varying fraction defective of the incoming lots. A comprehensive study is conducted to compare the performance characteristics of the optimum rectifying inspection plans with several other plans. The results show that the optimum plans usually produce the best outcome and always satisfy the design specifications. Moreover, the results of a sensitivity study indicate that the optimum plans are fairly robust. Finally, a manufacturing example illustrates the applications and the managerial significance of the optimum rectifying inspection plans.     

The economic effect of measurement error on variables acceptance sampling

Considerable attention to the economics of variables acceptance sampling has recently emphasized that the alpha-beta approach to sampling plan design may be far from optimal. Similarly, recent results conclusively show that measurement error manifested as bias (inaccuracy) and imprecision (variability) can have disastrous effects upon the OC curve of a variables sampling plan. This paper quantitatively combines economics and error and clearly illustrates the adverse monetary effects of imperfect measurement in situations where variables acceptance sampling is used     

Multiple deferred state sampling inspection

Multiple deferred state sampling is on attribute inspection procedure in which the decision criterion for each lot dictates one of three decisions: (1) accept the lot; (2) reject the lot; or (3) conditionally accept or reject the lot based on the disposition of future related lots. It is intended to supplement existing sampling procedures such as chain sampling, dependent stage sampling, exponential smoothed sampling, and fixed deferred state sampling.     

An improvement of variable sampling plans based on Kullback-Leibler information

A new and convenient design procedure for single sampling plans by variables is given by Arizono and Ohta (1986) based on the Kullback-Leibler information. In this paper an improvement of the original design procedure is presented and the revised procedure is compared with the bayesian procedure.     

科学应用抽样检验标准

自关于抽样检验的国家标准ＧＢ２８２８～２８２９—８１发布实施后,目前已发布了２０多个有关抽样检验的国家标准,但科学地、正确地选用抽样检验标准,在实际运用中还存在一些问题。产品质量监督抽样检验标准的选用是其中典型的事例。根据有关法规赋予产品质量监督检验...     

Transformation of analog-converter error during sampling

Translated from Izmeritel'naya Tekhnika, No. 1, pp. 5–6, January, 1988.     

Comparison of analytical error and sampling error for contaminated soil

Investigation of soil from contaminated sites requires several sample handling steps that, most likely, will induce uncertainties in the sample. The theory of sampling describes seven sampling errors that can be calculated, estimated or discussed in order to get an idea of the size of the sampling uncertainties. With the aim of comparing the size of the analytical error to the total sampling error, these seven errors were applied, estimated and discussed, to a case study of a contaminated site. The manageable errors were summarized, showing a range of three orders of magnitudes between the examples. The comparisons show that the quotient between the total sampling error and the analytical error is larger than 20 in most calculation examples. Exceptions were samples taken in hot spots, where some components of the total sampling error get small and the analytical error gets large in comparison. Low concentration of contaminant, small extracted sample size and large particles in the sample contribute to the extent of uncertainty.     

产品表面检测的变步长采样机制研究

针对产品表面缺陷检测系统中,由于定步长数据采集造成数据冗余量大、工作效率低的现象,提出变步长自适应的数据采集方式,以信号变化为采样依据实现调频。首先,基于采样步长的影响因素(乃奎斯特采样定理、识别尺寸、弧度畸变及空间畸变),推导出采样步长的函数关系表达式。其次,采用Matlab仿真平台,利用最小二乘法拟合,具体分析并验证被检产品各待检区域通过公式计算得到的采样步长均在95%的置信区间内。实验表明:推导出的采样步长公式为产品各待检区域旋转步长的合理选取提供理论依据;该模式可以根据不同信号实现频率自适应调节,可提高获取信息的利用率并减少数据冗余。     

数字采样测量噪声误差的估计和分析

针对高频数字采样中的附加噪声和时间抖动误差估计问题，给出了高频数字采样测量噪声误差的最大似然估计方法。在获得了相应的数学模型后，对最小二乘估计和最大似然估计方法进行了比较。仿真结果表明，对于此类误差最大似然估计比最小二乘估计有效性更好。另外最大似然估计的一个优点是用于不确定度计算的克拉美一罗下限值更易于获得。     

GPS单点定位误差分析

采用GPS单点定位技术与PPP解算软件,研究在静态GPS单点定位实践过程中,如何处理观测时间与观测误差的关系,以更好地提高数据采集准确度,为实际工作提供帮助。     

Economic acceptance sampling plans in complex multi-stage production systems

Specification of economic acceptance sampling schemes in complex multi-stage production systems is considered. The analysis is based on approximations of the distribution of lot fractions defective at all production stations of the system. A heuristic algorithm is developed that combines individual optimal solutions to subproblems corresponding to the production stations, to arrive at a near-optimal solution for the original problem. Numerical examples illustrate the application of the method to serial and non-serial manufacturing-assembly processes.     

Automatic sampling for CMM inspection planning of free-form surfaces

Coordinate measuring machines (CMMs) are used to examine the conformity of the produced parts with the designer's intent. The inspection of free-form surfaces is a difficult process due to their complexity and irregularity. Many tasks are performed to ensure a reliable and efficient inspection using CMMs. Sampling is an essential and vital step in inspection planning. Efficient and reliable approaches to determine the locations of the points to be sampled from free-form surfaces using the CMM were developed. Four heuristic algorithms for sampling based on the NURBS features of free-form surfaces are presented. The sampling criteria are equiparametric, surface patch size and the surface patch mean curvature. An algorithm for automatic selection of sampling algorithms performs complexity checks on NURBS surfaces, including the surface curvature changes and surface patch size changes, and selects the suitable sampling algorithm. Extensive simulations were performed using the developed methodologies to evaluate their performance using free-form surfaces with different degrees of complexity and compared with the uniform sampling pattern. The CMM measurement errors and manufacturing form errors have been simulated in these studies. The developed algorithms provide a useful tool in selecting the effective sampling plans for the tactile CMM inspection planning of free-form surfaces.     

浅谈家电产品监督抽查工作中的抽样环节

本文对监督抽查工作的定义及抽样要求进行介绍,通过对抽样环节中发现的问题进行分析,提出在新形势下做好抽样工作的建议。     

An alternative approach to accept on zero and accept on one sampling plans

Accept on zero and accept on one sampling plans are two common approaches to determine if a manufacturing lot can be accepted or rejected. Accept on zero plans inspect fewer items than the accept on one plans. On the other hand, accept on one plans have a higher probability of accepting a lot when the defect rate is between the Acceptable and Rejectable Quality Levels. This article proposes a double sampling plan whose probability of accepting a lot resembles an accept on one plan yet inspects considerably fewer items on average.     

Correctable sampling plans for increasing reliability of statistical quality control

We consider joint use, under conditions of statistical indeterminacy, of the results of sampling using an alternative test for preliminary information and an increase in the information obtained from experimental data.Translated from Izmeritel'naya Tekhnika, No. 12, pp. 14–16, December, 1994.     

Designing two mixed quick switching sampling plans for linear profiles

This study compares two proposed mixed quick switching sampling (QSS) plans for linear profiles as the quality characteristic. For the QSS plans, we recommend a binomial attribute plan for normal inspection and then a variable sampling plan for tightened inspection based on capability index C_puA of linear profiles with one-sided specifications. The difference between the two proposed QSS plans is in the tightened inspection. Tightened inspection of the first proposed plan is a single sampling using C_puA index, but tightened inspection of the second plan is a multiple dependent state repetitive (MDSR) plan based on C_puA index. The optimal parameters are obtained by nonlinear optimization. Simulation study for selecting parameters is conducted with various combinations of specified acceptable quality level (AQL), limited quality level (LQL), producer's risk, and consumer's risk. Simulation results confirm that the second proposed QSS plan which applies variable MDSR at tightened inspection performs better than another proposed plan. Hence, the approach of the second proposed plan is demonstrated in an illustrative example.     

The effects of timing jitter in sampling systems

Timing jitter generally causes a bias (systematic error) in the amplitude estimates of sampled waveforms. Equations are developed for computing the bias in both the time and frequency domains. Two principle estimators are considered: the sample mean and the so-called Markov estimator used in some equivalent-time sampling systems. Examples are given using both real and simulated data. It is shown that the bias that results from using the sample mean as an estimator can be approximated in the frequency domain by a simple filter function. The Markov estimator is shown to asymptotically converge to the population median. It is therefore an unbiased estimator for monotonic waveforms sampled with jitter distributions having a median of zero