Quality control

DR: Khaled seif

Programe: operations managment
Prepared by : ahmed nasr sayed
Group: 2nd w
Reg. no : 22126630
:Introduction
Quality control : A process that evaluates output relative to a standard and takes
Corrective action when output doesn’t meet standards.
Quality assurance that relies primarily on inspection of lots (batches) of previously
produced items is referred to as acceptance sampling.It is described in the chapter
supplement. Quality control efforts that occur during production are referred to as
statistical process control, and these we examine in the following sections. The best
companies emphasize designing quality into the process, thereby greatly reducing the
need for inspection or control efforts. As you might expect, different business
organizations are in different stages of this evolutionary process: Some rely heavily
on inspection. However, inspection alone is generally not sufficient to achieve a
reasonable level of quality. Many occupy a middle ground that involves some
inspection and a great deal of process control.
Inspection is a crucial aspect of quality control, comparing goods or services to a
standard. It is essential in lean organizations, where workers have responsibility for
quality. Inspection can occur before production, during production, and after
production, with acceptance sampling procedures and process control monitoring.
The main issues include the amount and frequency of inspection, the appropriate
points in the process, the location, and whether to inspect attributes or variables.
Inspections are essential in the production process, but they are not generally cost-
effective at every point. Managers must decide whether to remove computers from
the line and take them to a lab or test them where they are being made. The amount
of inspection can range from no inspection to multiple times. Low-cost, high-volume
items like paper clips, roofing nails, and wooden pencils require little inspection due
to low costs and reliable processes. High-cost, low-volume items with large costs
require more intensive inspections, such as critical components in manned-flight
space vehicles. Most quality control applications require some inspection, but it is
neither possible nor economically feasible to critically examine every part of a
product or service for control purposes. The cost of inspection, interruptions, and
testing methods typically outweigh the benefits of 100% inspection. Inspection is a
process subject to variation, and boredom and fatigue can cause inspection mistakes.
The amount of inspection needed is governed by the costs of inspection and the
expected costs of passing defective items. It may not be cost-effective to catch every
defect, especially if the cost of inspection exceeds the penalties associated with
letting some defects pass. Current thinking suggests that every reduction in defective
output reduces costs, although not primarily by inspection. High-human involvement
operations require more inspection effort than mechanical ones. The frequency of
inspection depends on the rate of process out-of-control or the number of lots being
inspected. Computerized automatic inspections at critical points are cost-effective for
high-volume, repetitive operations. Inspecting manufacturing operations involves
focusing on various points to ensure quality standards. These include raw materials,
finished products, costly operations, irreversible processes, and covering processes.
Restricting inspection efforts to these areas can reduce costs and improve customer
satisfaction. Inspection improves process yield by reducing scrap rate and reducing
inspection needs at key points. Centralized inspections are preferred for on-site
situations, while lab tests are preferred for specialized tests. On-site inspections offer
quicker decisions and avoid extraneous factors. Labs offer specialized equipment,
favorable test environment, and self-inspections for errors. Quality control ensures
process conformance and corrective action based on output variations. Process
variability involves output variations that are within acceptable ranges. To improve a
stable process, it's crucial to determine if these variations are random or inherent.
Common variability, caused by minor factors, differs from process to process.
Machines have higher natural variability due to worn parts and design improvements.
Assignable variation, or non-random variation, can be identified and eliminated.
Sample statistics, such as the mean or number of occurrences, can be used to judge
the randomness of process variations. The central limit theorem states that as sample
size increases, sample averages approach normal distribution, even for small sample
sizes. The normal distribution approximates the actual sampling distribution, with
larger samples having narrower distributions.
The control process involves defining and correcting actions based on specific
characteristics, such as paint thickness, hardness, and resistance. Control involves
measuring characteristics, comparing them to a standard, evaluating out-of-control,
correcting nonrandom variability, and monitoring results to ensure corrective action
is effective. Corrective action involves identifying causes, correcting deviations, and
following up to ensure problems are resolved.
Control charts, developed by Walter Shewhart, are crucial tools in statistical process
control. They plot sample statistics to distinguish between random and nonrandom
variability, with upper and lower limits defining acceptable variation. Control charts
monitor process output to ensure randomness, with a normal sampling distribution
being a minor challenge.
Control limits the lines dividing random and nonrandom deviations from the
distribution mean. Type I error indicates a process is not in control, while Type II
error indicates it is in control. Four control charts are used for variables and
attributes, counting and measuring data on a continuous scale. Variables data is
measured, while attributes data is measured on a specific scale.
Mean and range charts are used to monitor variables, with mean charts focusing on
the central tendency of a process and range charts monitoring the dispersion of a
process. Mean control charts, also known as ¯x charts, are based on a normal
distribution and can be constructed using upper control limit (UCL) and lower
control limit (LCL) formulas. The formulas involve calculating the standard
deviation of a process, n, and the number of standard deviations that control limits
are based on, and x is the average of sample means. Range control charts (R-charts)
monitor process dispersion and are sensitive to changes in dispersion. They are
similar to mean charts but with different sampling distributions. Control limits for
range charts are found using the average sample range and formulas like UCL = D4
and LCL = D3.
Mean and range control charts offer different perspectives on a process, with mean
charts sensitive to shifts in the process and mean and range charts sensitive to
changes in process dispersion. Both provide more complete information than either
chart alone. However, single charts may suffice in some cases.
Attribute control charts are used to count the number of defective items in a sample,
rather than measuring them. There are two types: p-charts for the fraction of
defective items in a sample and c-charts for the number of defects per unit. P-charts
monitor the proportion of defective items generated by a process, with the binomial
distribution as the theoretical basis.
A c-chart controls defect occurrences per unit using the Poisson distribution, with
control limits UCLc = c + z√____c and LCLc = c − z√_____c
Control charts are essential for organizations to monitor and control processes, but
they can add cost and time to the process. Managers must make important decisions
about the use of control charts, such as selecting the appropriate points in the
process, selecting the appropriate sample size, choosing the control chart type
(variables or attributes), and determining the frequency of sampling. The sample size
is crucial for revealing changes in the process, as larger samples may result in more
lost products and longer process delays. Managers can choose between using a mean
chart for variables or a p-chart for attributes, weighing the time and cost of sampling
against the information provided. Analysts often use run tests to detect abnormalities
in a process and correct out-of-control processes. Run tests examine patterns in a
sequence of observations, such as trends, cycles, or bias, to identify assignable
causes of variation. Two widely used run tests are up and down and above and below
the median. To identify patterns in control chart data, transform the data into As and
Bs and Us and Ds, count the number of runs in each case, and compare it with the
expected number of runs in a completely random series. The expected number of
runs is a function of the number of observations in the series. Chance variability is
measured by the standard deviation of runs. To differentiate between chance
variability and patterns, use sampling distributions for median and up/down runs.
Ideally, both control charts and run tests should be used to analyze process output,
along with a plot of data. The procedure involves three steps: calculating control
limits, conducting median and up/down run tests, and checking for patterns. If no
indication is found, the process output is likely not random. If both test scores are
within z = ± 2.00, proceed to Step 3. If no pattern is found, conclude the output is
random and the process is in control.
Nonrandom variation occurs when a process observes a point outside control limits
or a large z-value. Managers should have response plans in place to investigate the
cause, which may be a false alarm or a real indication of an assignable cause. If a
cause is found, address it. If a good result is found, change the process to achieve
similar results on an ongoing basis. In more complex cases, operators can be trained
to handle simple problems, while teams may be needed to handle more complex
problems. Problem-solving often requires using various tools to find the root cause.
Once the root cause is found, changes can be made to reduce the chance of
recurrence.
Process capability is crucial after establishing stability and avoiding nonrandom
variations. The analysis focuses on determining if a process can produce acceptable
output within acceptable ranges. Three terms describe variability in process output,
each relating to a slightly different aspect. Specifications or tolerances are
established by engineering design or customer requirements, indicating a range of
values for individual units of output to be acceptable.
Control limits are statistical limits that reflect the variability in sample statistics due
to randomness. Process variability, on the other hand, is the natural or inherent
variability in a process, measured in terms of the process standard deviation. Control
limits are based on sampling variability while sampling variability is a function of
process variability. However, there is no direct link between specifications and
control limits or process variability. Specifications are specified in terms of the
output of a product or service, not the process by which it is generated. Therefore,
the output of a process may or may not conform to specifications, even if the process
is statistically in control. It is crucial to consider the capability of a process, which
refers to the inherent variability of process output relative to the variation.
Capability analysis is a crucial process management process that assesses a process's
ability to meet design specifications for its intended use. It is performed on a process
that exhibits random variation to determine if the range of variation is within the
specifications. If the process is capable, it is considered "capable." However, if it is
not, the manager must correct the situation. three cases are illustrated: the first case,
where process capability and output specifications are well matched, where nearly all
output can meet specifications. The second case is where process variability is less
than expected, where almost 100% of output is well within tolerance. The third case
is where specifications are tighter than the process's capabilities, where a significant
percentage of output fails to meet specifications. To ensure a process is in control, it
is essential to check its ability to meet specifications, rather than simply setting up a
control chart to monitor it. This approach helps ensure that the process is capable of
producing acceptable output, ensuring its effectiveness in achieving desired output.
A manager should ensure the process is in control and within specifications before
production begins. In cases like case C, a manager can consider redesigning the
process, using an alternative, retaining the current process but eliminating
unacceptable output using 100% inspection, and examining specifications to
determine if they are necessary or could be relaxed without negatively affecting
customer satisfaction. Process variability is the key factor in process capability,
measured in terms of the process standard deviation. To determine if a process is
capable, compare ± 3 standard deviations to the specifications, such as a 10-minute
service with an acceptable range of variation of ± 1 minute.
Enhancing process capability involves reducing inherent variability, simplifying,
standardizing, making processes mistake-proof, upgrading equipment, or automating
to improve efficiency and effectiveness. Improved process capability reduces
inspections, lowers warranty costs, reduces service complaints, and increases
productivity while limiting customer control limits.
Taguchi, a Japanese quality expert, believes that deviation from the target value
represents poor quality, and the greater the deviation, the higher the cost. He suggests
that reducing variation in a process and increasing its capability ratio, can lower the
cost of poor quality and reduce the loss to society.
Capability indexes have limitations, including potential instability, incorrect
inferences about unacceptable output distribution, and the use of the Cp index, which
may give misleading results when the process is not centered. These risks make it
crucial to carefully consider and use a capability index when assessing a process's
stability, distribution, and overall performance.
Quality is crucial for customer satisfaction, making it crucial for business
organizations to achieve and maintain high standards. Quality assurance and product
and service design are vital links in the process. Organizations should continuously
seek to increase the capability of their processes, moving from inspection and control
charts to a built-in approach where quality is built into products and processes.
Opportunities for continuous process improvement should be seen in processes with
nonrandomness or inability.