Quality in Production

• Process Control in Quality Management Systems involves proactively monitoring and controlling processes so that they produce products with desirable characteristics consistently.
• Process Control is the most important part of the quality effort during manufacturing.
• Process Parameters are used in quality literature to indicate both the variables of a process that influence the product characteristics, as well as the quantities that define the distribution of a process.
• Maintenance of the process parameters at the chosen levels is accomplished using the control charts.
• Control Charts were proposed by Dr. Walter Shewhart, working in the early 20s at the Bell Telephone Laboratories in Princeton, New Jersey.
• The variability in a process parameter, or product characteristic, can be from two sources: Stable system of chance causes and Assignable Causes.
• A typical Control Chart has a centerline (CL) and two control limits, an upper control limit (UCL) and a lower control limit (LCL), representing the limits for chance-cause variability and are computed using data drawn from the process.
• If the values from all the samples taken during a period of time lie within the limits, the process is said to be “in control” during that time period.
• If the value from any of the sample plots lie outside either one of the limits, the process is said to be “not in control” during that time period.
• Proper check sheets or standard forms must be developed for recording relevant process information and for recording and analyzing data.
• The run chart can be used with measurements such as diameter, weight of parcels, waiting time in the doctor’s office, etc., or attributes such as number of defects on a casting, number of patients waiting for appointment, or number of C-sections performed in a month at a clinic.
• According to this rule, the control limits will be placed at three standard deviations of the statistic being plotted, from the centerline.
• The run chart can be used with both measurements and attributes.
• The recommended sample size for X and R-charts is four or five.
• A more practical approach is to take samples more frequently during the initial stages of controlling a process and then reduce the frequency once stability is attained.
• The control chart is also subject to Type I and Type II errors.
• There are no limit lines on the run chart.
• Rational subgrouping is an important idea that should be understood and employed correctly in order to get the most out of control charts.
• When the process is in-control, it does not mean that the process is also capable.
• The S-chart, or the standard deviation chart, is used when the sample size must be larger than 10 because that extra sensitivity is needed for the X-chart, the R-chart cannot be used because of the poor efficiency (i.e., larger variability) of the statistic R when the sample size is large.
• Many advantages are realized when processes are controlled using control charts, also referred to as “statistical process control” (SPC) tools.
• Two types of data are used in control charts: Measurement Data and Attribute Data.
• A capable process produces “all” units of its output within specification.
• The control limits for Attribute Control Charts are calculated using the formula: P- Chart.
• If standards are given, Moving average and moving range charts are used.
• Attribute Control Charts are used to control and minimize the proportion of defects in a process with a sample size of n ≥ 20.
• The Standarized DNOM Chart plots the number of standard deviations each sample average is from the nominal value.
• Benefits of using control charts include avoidance of defectives, improved customer relationships, increased profitability, improved worker morale, better knowledge of processes and their capabilities, increased market share, and better knowledge of processes and their capabilities.
• Capability of a process with a measurable output has normal distribution as it produced a measurable output.
• A process in control may not fully be capable of producing products that meets customers satisfaction.
• If the process produces its output from a former source, the average X and the standard deviation S will estimate μ and σ, respectively.
• The process has changed if more than seven consecutive plots fall above or below the center line or more than seven plots are in a run-up or run-down.
• If the process produces its output from the latter source, X will estimate μ, and R /d2 will estimate σ, where R is the CL of the R-chart and d2 is the correction factor that makes the R an unbiased estimator for σ.
• The control chart for individuals, or the X-chart, is normally used along with a chart for successive differences, which is known as a moving range chart, or MR chart, with subgroup size n = 2.
• The process is not in control if there is a run of six or more points above or below the median, a run up or run down of five or more points, too many or too few runs above or below the median as determined by comparing with tabled critical values, or an astronomical point (outlier plot).
• If the S-chart is used instead of the R-chart, SI /c4 will provide an estimate for σ.
• The DNOM Chart is applied to small processes.
• Measurement Control Charts are designed to control measurements, with the assumption that those measurements follow the normal distribution.
• When a measurement is to be controlled, both its average and its variability must be controlled.
• The most popular combination used to control a measurement is that of X-chart and R-chart, the former to control the process mean and the latter to control the process variability.