Production data collection

Data Collection—the previously introduced control components require production data to be recorded with high quality. Therefore, also production data collection models need to be individualized for each product. 

The check sheet shown below, which is tool number five, is a simple technique for recording data (47). A check sheet can present the data as a histogram when results are tabulated as a frequency distribution, or a mn chart when the data are plotted vs time. The advantage of this approach to data collection is the abiUty to rapidly accumulate and analy2e data for trends. A check sheet for causes of off-standard polymer production might be as follows ... 

Data Collection and Analysis Pertinent to the PA s Development of Guidelines for Procurement of Highway Construction Products Containing Recovered Materials, EPA Contract 68-01-6014, Draft, Vol. 1, Issues and Technical Summary, Franklin Associates, Ltd., and Valley Forge Laboratory, Inc., July 6, 1981. 

With the advent of the microprocessor, digital technology began to be used for data collection, feedback control, and aU other information processing requirements in production facUities. Such systems must acquire data from a variety of measurement devices, and control systems must drive final actuators. 

Data collected by modern analytical instalments are usually presented by the multidimensional arrays. To perform the detection/identification of the supposed component or to verify the authenticity of a product, it is necessary to estimate the similarity of the analyte to the reference. The similarity is commonly estimated with the use of the distance between the multidimensional arrays corresponding to the compared objects. To exclude within the limits of the possible the influence of the random errors and the nonreproductivity of the experimental conditions and to make the comparison of samples more robust, it is possible to handle the arrays with the use of the fuzzy set theory apparatus. 

Decisions affecting the future direction of the organization and its products and services are made from information gleaned through market research. Should this information be grossly inaccurate, over optimistic or pessimistic the result may well be the loss of many customers to the competition. It is therefore vital that objective data is used to make these decisions. The data can be primary data (data collected for the first time during a market research study) or secondary data (previously collected data). However, you need to be cautious with secondary data, as it could be obsolete or have been collected on a different basis than needed for the present study. 

This requirement is similar to that in clause 4.14.3 under Preventive action since the data collected for preventive action serves a similar purpose. In one case an analysis of company-level data serves to identify overall trends and predict potential failures that will affect achievement of the goals. In the preventive action case, the data serves to identify local and overall trends and predict potential failures that will affect achievement of specified requirements for the product, process, and quality system. It would be sensible to develop a data collection and analysis system that serves all levels in the organization, with criteria at each level for reporting data upwards as necessary. You should not treat this requirement separately from that for preventive action since the same data should be used. However, the explanation given in clause 4.1.5 of Operational performance does include some factors that may not be addressed in your preventive action procedures. 

The nonconformity data should be collected and quantified using one of the seven quality tools (see Part 2 Chapter 14), preferably the Pareto analysis. You can then devise a plan to reduce the 20% of causes that account for 80% of the nonconformities. However, take care not to degrade other processes by your actions (see Theorg ofcon-staints in Part 2 Chapter 2). The plan should detail the action to be taken to eliminate the cause and the date by which a specified reduction is to be achieved. You should also monitor the reduction. The appropriate data collection measures therefore need to be in place to gather the data at a rate commensurate with the production schedule. Monthly analysis may be too infrequent analysis by shift may be more appropriate. 

The fact that the model connecting error types with their causes may change as a result of gaining further experience with the data collection system means that the informahon gathered on the PIFs in a situation may also change. For example, if incident data indicates the neglect of safety procedures because of production pressures, then the questions relating to this area wUl need to be extended. 

Control laboratories in the canned food industry are usually divorced from the research organization to a lesser degree than is the case in the chemical and allied industries. For this reason, a closer relationship exists between the problems of the control laboratory and the research laboratory. Although from a research standpoint this condition is often considered undesirable, it has considerable merit in the case of the canned food industry, in which production may be seasonal and often of rather short duration. The collection of control data in many instances may also serve for research purposes—for example, in the case of soil analyses, which may be correlated with agricultural research designed to improve crop yields. Because the variables which affect the quality of canned foods must usually be investigated rather extensively, and often over a period of more than one year, the application of statistical methods to data collected for control purposes can conceivably make a substantial contribution to a research program. 

Also the a-ester sulfonates are less important today. In the Federal Republic of Germany, for example, the total production of surfactants was about 700,000 t/a in 1993. For a more detailed analysis of different types of surfactants, use must be made of data collected before the unification of Germany. In 1988 the consumption of surfactants in detergents was about 227,500 t/a, the consumption of anionic surfactants was about 116,000 t/a and less than 1000 t/a of a-sulfo fatty acid esters [5] (the values refer to German Detergent Law). 

Products in Group 3 seem to us to represent the future of practical batch process control. In such systems, modern workstations perform the single-user functions (e.g control system design, set-up, and maintenance operator interface data collection historical reporting) for which they were designed, while powerful multitasking controllers perform actual control. As computer hardware and software standards continue to evolve toward distributed networks of processors optimized for specific kinds of tasks, such systems will, we feel, proliferate rapidly. 

There has been a push for direct data collection (DDC) as an alternative to remote data capture (RDC). In this approach most of the required clinical data are acquired directly from existing patient record systems such as MRI machines, ECG, EEG, TTM, laboratories, and other measurement equipment. This approach eliminates the need for paper transcription and reentry to another system. It promises error-free and resource-efficient data capture, which allows early locking of the database and therefore potentially earlier product launch [30]. 

O3 + terpene products Rate =. [03] [terpene] We expect the reaction rate to depend on two concentrations rather than one, but we can isolate one concentration variable by making the initial concentration of one reactant much smaller than the initial concentration of the other. Data collected under these conditions can then be analyzed using Equations and, which relate concentration to time. For example, an experiment could be performed on the reaction of ozone with isoprene with the following initial concentrations ... 

To construct the reference model, the interpretation system required routine process data collected over a period of several months. Cross-validation was applied to detect and remove outliers. Only data corresponding to normal process operations (that is, when top-grade product is made) were used in the model development. As stated earlier, the system ultimately involved two analysis approaches, both reduced-order models that capture dominant directions of variability in the data. A PLS analysis using two loadings explained about 60% of the variance in the measurements. A subsequent PCA analysis on the residuals showed that five principal components explain 90% of the residual variability. 

---