Process monitoring tools control charts

The control chart is the basic analytical tool of SPC and is used for first assessing a process, then for monitoring a process output with respect to on-target control and process variability. A control chart is basically a time plot of a statistic calculated from a variable associated with a process. This variable may either be a process variable, such as temperature or flow rate, or a product variable, such as fill weight or potency. Examples of statistics are an individual measurement, an average of two or more measurements, a percentage of defective output items, a count of defect occurrences in time or space, or a measure of variation such as a range or standard deviation of two or more measurements. 

Statistical process control (SPC), also called statistical quality control and process validation (PV), represents two sides of the same coin. SPC comprises the various mathematical tools (histogram, scatter diagram run chart, and control chart) used to monitor a manufacturing process and to keep it within in-process and final product specification limits. Lord Kelvin once said, When you can measure what you are speaking about and express it in numbers, then you know something about it. Such a thought provides the necessary link between the two concepts. Thus, SPC represents the tools to be used, while PV represents the procedural environment in which those tools are used. 

Probably the two most basic generic industrial problems commonly approached using statistical methods are those of (1) monitoring and maintaining the stability/consistency of a process and (2) assessing the capability of a stable process. This section provides a brief introduction to the use of tools of control charting in these enterprises. 

Control charts are an excellent analysis tool to both monitor and improve in-process performance during process development and later during production, where it is desired to follow process characteristics over time within batches or runs. The most common examples of tablet process characteristics that are measured in-process are weight, thickness, and hardness. The parameters measured need to be controllable so that adjustments can be made. During the initial runs, it is desirable to limit process adjustments to a minimum to observe the process in its natural state. Any adjustments made should be recorded and explained. Out-oflimit results should never be removed prior to performing a process capability analysis. If special cause variation is detected, then process improvements should be made to eliminate the special cause variation. 

Various types of control charts, a tool that is most often used for internal quality control purposes, are often based on values given by reference materials. For this purpose, in-house reference materials should be preferably used, as certified reference materials are too expensive to be used to monitor the stability of the measurement process. 

The tool that is used to monitor process variation over time is known as the control chart. Control charts originate from the work of Walter Shewhart (1927) and are often referred to as Shewhait control charts. Effectively, process observations based upon collected samples or subgroups, at fixed points in time, are plotted in accordance to time. As long as the current observation is within fixed... 

Control charts are powerful tools for monitoring the variation of a process. Furthermore, the nonnatural trends on a control chart can provide significant diagnostic information regarding the cause of a process disturbance. In this section, prevalent trends from out-of-control processes are presented and the suspect causes are briefly discussed. References such as AT T (1985) and Montgomery (1996) provide more in-depth presentations of this discussion. 

Many statistical tools have been developed to control critical process parameters. The most commonly used is the control chart, which is an effective way to monitor and control processes and can be defined for both vtuiables and attributes data. The selection of variables data will typically make basic statistical tools more efficient (i.e., lower sample size requirements to achieve necessary confidence levels). 

SPC helps in detecting, identifying, and eliminating unpredictable sources of variability in the process. Moreover, it helps monitoring the process by issuing signals whenever deviations from in-control conditions are detected. Statistical control charts are the basic tools to implement SPC in manufacturing processes. 

Normality assumption Most traditional SPC tools are based on the assumption that the process output characteristic is normally distributed, among which Shewhart control charts and multivariate control charts. In some cases, the central limit theorem can be used to justify approximate normality when monitoring means, but in numerous cases normality is an untenable assumption, and one is unwilling to use another parametric model. A number of nonparametric methods are available in these cases. As data availability increases, nonparametric methods seem especially useful in multivariate applications where most methods proposed thus far rely on normality. 

Control chart It is a tool in statistical process control to monitor the number of defects found in a product or a process overtime and stndy the variation and its source. 

From these values, a pair of control charts is created. These charts are used to plot the SPC data as it occurs. They are used as a visual tool to monitor the process. Chart 8.14 is an example of the X-Bar SPC chart that monitors a process. For this chart we will use X = 1.23 and R =. 45. 

The philosophy of SPC is to monitor the output of a process and determine when control action is necessary to correct deviations of the output from its setpoint. The most common tool for accomplishing this is the Shewhart (x-bar) chart shown in Figure 5.19. In the discrete parts manufacturing industries, multiple samples are taken at fixed intervals. Quality tests are run on these samples, and the mean is plotted on one Shewhart chart, and the range on another. In the absence of a disturbance, the means should be normally distributed around the setpoint. If the upper and lower control limits (UCL and LCL, respectively) are placed at three standard deviations above and below the target, a range is defined into which all of the means should fall. The... 

---