Statistical process control data collection

Statistics refers to the scientific methods applied to the collection, organization, interpretation, and presentation of information—numerical data. For statistical process control (SPC), data types are divided into attributes or variables. 

Statistical analysis such as Statistical Process Control (SPC) may be used to derive performance parameters as well as track and trend for alert/alarm conditions. Automated moifitoring tools may be available to assist in the collection of relevant data. A record of any such tools used should be maintained and any validation requirements considered. 

To accomplish this, we must collect data. But how we collect that data is as important as the data themselves. Some data are worthless, some are priceless. The conditions and procedures used to find data ultimately determine their value. Statistical quality control (SQC), statistical process control (SPC), total quality management (TQM), and six sigma are all passive approaches to data collection. These procedures only observe and report what is happening. They cannot find the analytical cause-and-effect relationships needed for true process understanding and for controlling the sources of variability. 

Well-known and documented techniques exist to monitor product variation while it is within the producer s environment. Most of the techniques require the observations or data to be statistically independent. That is, the data for a specific performance measurement are assumed to have no relationship to prior or successive observations. It is assumed that no correlation exists between data collected prior to or following a specific observation. The techniques used to monitor such data are collectively called statistical process control (SPC). These techniques are utilized in consumer-oriented industries. Some of the more prominent or useful techniques are presented in this chapter. Specifically, seven tools for SPC are reviewed, and their applicability is examined. Furthermore, common and improved approaches for process capability analysis are presented. 

In the Principles of Quality course, students use advanced statistics and mathematics to work with operational data. Process technicians collect, organize, and analyze data during routine operations. The statistical approach works well with statistical process control and control charts. A variety of processes can easily be adapted to fit these quality tools. Examples of these include equipment and quality variables process variables include pressure, temperature, flow, level, and analytical parameters. 

In a Principles of Quality course, process technicians collect, organize, and analyze data during routine operations, and study the background and application of quality concepts. Topics include team skills, quality tools, statistics, economics, and continuous improvement. The focus is on the application of statistics, statistical process control, math, and quality tools to process systems and operations. 

These technologies, when apphed properly, can reduce catastrophic failure, and thus maintenance cost. One other apphcation is statistical process control (SPC). This predictive tool can be used to predict failures, but a plan must be in place first, for the data collection process is critical. If a CMMS is used, then the proper system architecture must be developed, along with associated processes and procedures that allow for accurate data collection. 

Collecting data for statistical process control (SPC) to evaluate mean time between failures (MTBF), cost of maintenance, failure rates, and so on... 

Statistical process control results for all production lines were summarized at Table 1 below. Theoretical calculation of process capability indices from the collected data were very low, because the upper tolerance limit of polypropylene pipes indicated in Turkish Standards (TS 10595) is very high, so process capability indices could not represent the process actual performance... 

Decision Process. In many cases, the decision regarding the need for exposure reduction measures is obvious and no formal statistical procedure is necessary. However, as exposure criteria are lowered, and control becomes more difficult, close calls become more common, and a logical decision-making process is needed. A typical process is shown in Eigure 2. Even when decision making is easy it is useful to remember the process and the assumptions involved. Based on an evaluation, decisions are made regarding control. The evaluation and decision steps caimot be separated because the conduct of the evaluation, the strategy, measurement method, and data collection are all a part of the decision process. 

Data collected for each process phase may also be evaluated statistically to evaluate objectively whether a process change was better or worse than the preceding one. For example, through analysis of variance, it would be possible to determine whether each process phase had demonstrated continued process control or clear improvement. The revalidation approach would thus allow the QA (or production technical services) group to proactively manage its responsi-... 

The functions now realized by microprocessors include the control of the optical system (lamp and analytical wavelength selection), selection of the kind of data collected (e.g., absorbance, concentration), zero-adjustment, autocalibration and control of measurement parameters [21]. The microprocessor determines the equation of the regression curve and provides statistical processing of the results. It can also be programmed to measure the absorbance, the % transmittance at a selected wavelength, or the concentration based on the relationship (linear or non-linear) established between the measured absorbance and the concentration. 

First, it is the assumption of most control chart apphcations that the process from which observations are collected is stable. That is, the statistical behavior of the process is time invariant in that the underlying distribution is fixed, yielding a fixed mean and variance. In some applications, process observations are collected such that multiple processes may feed the data-coUection station. As a result, the observations may have a tendency to cluster around the control limits and be sparsely observed around the CL or the mean. This observation is commonly referred to as a mixture. 

Prior to the widespread implementation of supervisory control and data acquisition (SCADA) and human-machine interface (HMI) systems, most SPC and SQC was performed by quality-control departments as an off-line process. Data was collected from test stations, laboratories, etc. and statistical analysis was performed later. SCADA/HMI systems, however, have made it feasible to provide plant-floor SPC charts using data collected in real time directly from the process. Fabricators that want to standardize SPC and SQC to increase their use find they need the two following functions (1) provide the plant floor with SPC charts and (2) make data collected by SCADA systems available for off-line analysis. Available is SPC and SQC software to support these efforts. Recognize that the bulk of SPC s value is derived from process improvements developed from offline SQC analysis. 

---