Statistics acceptance sampling

Thermal comfort may be defined as "that condition of mind in which satisfaction is expressed with the thermal environment" (4). It is thus defined by a statistically vaUd sample of people under very specific and controlled conditions. No single environment is satisfactory for everybody, even if all wear identical clothing and perform the same activity. The comfort zone specified in ASHRAE Standard 55 (5) is based on 90% acceptance, or 10% dissatisfied. 

Figure 6. Decision contours resulting from OSHA Compliance Criteria AL = 0.835, and UAL = 1.165, and assuming that the decision is based on six statistically independent samples. In comparison with Figure 5, many more environments are subject to a citation. This even includes some environments within the acceptable region bounded by the A EL contour in Figure 2.

Let us illustrate the benefits of higher order on a concrete analytical example measurements of concentration of Mg2+ with an ISE and with an optical sensor. After linearization of the potentiometric signal, the two experiments can be displayed as a bilinear plot (Fig. 10.2). Contained in this plot is an unusual sample point S, which clearly falls out of the linear correlation because it lies outside the statistically acceptable 3a noise level. This outlier is an indication of the presence of an interferant. Its presence is clearly identified in this bilinear plot from combined ISE and optical measurement, although it would be undetected in a first-order sensor alone. 

Survey results need to be very carefully assessed before they can be accepted and this must take account of how the analytical methods were applied, as well as the representativeness of the samples of food of the general supply. Since analysis for PCDDs, PCDFs and PCBs in food is a particularly complex and hence expensive and time-consuming process, particular care needs to be taken that statistically sufficient samples are analysed in each survey. 

LD Denotes the lethal dose that kills 50% of a sample of experimental animals, and as such is a statistically acceptable measure of acute toxicity, though not necessarily a very meaningful pharmacological or clinical measure. 

The details of the assessment of stability data are under intense discussion within the scientific community. A majority of laboratories evaluate data with acceptance criteria relative to the nominal concentration of the spiked sample. The rationale for this is that it is not feasible to introduce more stringent criteria for stability evaluations than that of the assay acceptance criterion. Another common approach is to compare data against a baseline concentration (or day zero concentration) of a bulk preparation of stability samples established by repeated analysis, either during the accuracy and precision evaluations, or by other means. This evaluation then eliminates any systematic errors that may have occurred in the preparation of the stability samples. A more statistically acceptable method of stability data evaluations would be to use confidence intervals or perform trend analysis on the data [24]. In this case, when the observed concentration or response of the stability sample is beyond the lower confidence interval (as set a priori), the data indicate a lack of analyte stability under the conditions evaluated. 

A quality control chart is a time plot of a measured quantity that is assumed to be constant (with a Gaussian distribution) for the purpose of ascertaining that the measurement remains within a statistically acceptable range. It may be a day-to-day plot of the measured value of a standard that is run intermittently with samples. The control chart consists of a central line representing the known or assumed value of the control and either one or two pairs of limit lines, the inner and outer control limits. Usually the standard deviation of the procedure is known (a good estimate of cr), and this is used to establish the control limits. 

OC curves for standard acceptance-sampling plans are derived under the assumption that the quality of items can be modeled as independent and identically distributed (i.i.d.) Bernoulli random variables. Although this model is often plausible, the quality of items produced by some processes exhibit statistical dependence. The goal of this simulation experiment is to estimate the OC curve for sampling plan (10, 1) when item quality is dependent. 

Note Although the terms statistical process control (SPC) and statistical quality control (SQC) are often used interchangeably, there are various differences between these terms. SQC is a broader concept including descriptive statistical methods, acceptance sampling, and SPC as commonly adopted tools. Ishikawa (Ishikawa 1976) points out that statistical process control and statistical quality control use the same set of tools to control respectively the input of a process (independent variables) and the output of the process (dependent variables). Other SPC/ SQC advocates further elaborate this concept by differentiating these terms according to the type of data elaborated by the tools SPC is based on process signal data analysis, while SQC is based on product feature-related data. 

Weigand (1994) reported the point-count results of the ASTM microscopy task group on one of the Standard Reference Clinkers supplied by the National Institute of Standards and Technology (Gaithersburg, Maryland). Compared to the Bogue calculations, the microscopical data are roughly 5% higher for alite, 5% lower for belite, 4% lower for C A, and 0.5% lower for ferrite. It was further shown that 3000 points per sample would provide sufficient data for statistical acceptance of port-land cement clinker polished sections. 

This chapter discusses several statistical principles that are used in pharmaceutical quality decisions, such as normal distribution, rounding, confidence interval, standard deviation, outliers, operating characteristic curves, acceptance sampling. Examples have been embedded in a pharmaceutical context. 

The system of statistical end control, as mentioned in the previous section is called Acceptance Sampling. Elements are diverse parameters such as AQL, Producer s Risk, LQL, Consumer s Risk, the method of inspection or analytical procedure and the OC curve derived from them. Typically a plan contains not only the maximum and minimum limits of the content of product or batches but also the relative frequencies by which the outcome on content may be passed (or not) and what should be done accept or reject. 

Unfortunately this type of acceptance sampling is not always applied. One reason is without doubt that just stating the content alone as requirement is not sufficient. A detailed procedure for the assay method is required from which the margins follow or a tolerance interval is given from which the details of the assay method can be derived. Mixing up of these two approaches leads to an unclear situation. In the following subsections the second approach will be followed as it is mostly used in the Statistical Quality Control (SQC) and Statistical process control (SPC) community. 

There are two main facets of statistical quality control. One of them is the use of process control charts for in-process manufacturing operations. These charts, also referred to as variables control charts or attributes control charts, are aimed at evaluating present as well as future performance. The other facet of statistical quality control is acceptance inspection or acceptance sampling. This technique forms the basis for scientihcally evaluating past performance and accepting or rejecting the product. 

Acceptance sampling is a widely used and accepted statistical quality control technique. The acceptance sampling technique calls for selecting a sample randomly from a lot and deciding whether to accept or reject the lot, based on the number of defective items found in the sample. 

The probabilistic nature of a confidence interval provides an opportunity to ask and answer questions comparing a sample s mean or variance to either the accepted values for its population or similar values obtained for other samples. For example, confidence intervals can be used to answer questions such as Does a newly developed method for the analysis of cholesterol in blood give results that are significantly different from those obtained when using a standard method or Is there a significant variation in the chemical composition of rainwater collected at different sites downwind from a coalburning utility plant In this section we introduce a general approach to the statistical analysis of data. Specific statistical methods of analysis are covered in Section 4F. 

If the null hypothesis is assumed to be true, say, in the case of a two-sided test, form 1, then the distribution of the test statistic t is known. Given a random sample, one can predict how far its sample value of t might be expected to deviate from zero (the midvalue of t) by chance alone. If the sample value oft does, in fact, deviate too far from zero, then this is defined to be sufficient evidence to refute the assumption of the null hypothesis. It is consequently rejected, and the converse or alternative hypothesis is accepted. 

If draws can be made from the posterior distribution for each component conditional on values for the others, i.e., fromp(Q,i y, 6,- J, then this conditional posterior distribution can be used as the proposal distribution. In this case, the probability in Eq. (23) is always 1, and all draws are accepted. This is referred to as Gibbs sampling and is the most common form of MCMC used in statistical analysis. 

The information obtained during the background search and from the source inspection will enable selection of the test procedure to be used. The choice will be based on the answers to several questions (1) What are the legal requirements For specific sources there may be only one acceptable method. (2) What range of accuracy is desirable Should the sample be collected by a procedure that is 5% accurate, or should a statistical technique be used on data from eight tests at 10% accuracy Costs of different test methods will certainly be a consideration here. (3) Which sampling and analytical methods are available that will give the required accuracy for the estimated concentration An Orsat gas analyzer with a sensitivity limit of 0.02% would not be chosen to sample carbon monoxide... 

---