U3 Flashcards
A firm can use many techniques to follow requirements or perform to what has been previously specified. This unit outlines seven elementary quality tools (“Q7”) used to?
monitor quality, e.g., on the production floor.
The unit also explains seven management tools that are?
used by managers to control quality (“M7”).
The unit ends with more complex quality techniques, which help
firms ?
capture, visualize, and analyze problems.
The term quality technique is rather broad and includes quality tools, management tools, and further techniques.
Many of the tools, especially the Q7 and M7 tools, are easy to use by?
assembly workers or line managers without in-depth knowledge of quality control or complex computer software.
The visualization tools can be implemented by using a standard spreadsheet program and can be?
adapted to different types of problems.
Many of these techniques have their origin in?
the 1960s
when Japanese firms wanted to increase the quality of their products.
A lot of companies were successful in finding and eliminating?
problems in the production process.
tools were kept as simple as possible?
Because many of the tools used were developed for workers in production plants.
Tools. They are also called “seven tools,” “seven new tools,” “quality control tools,” and “creativity techniques”?
-These were used to capture errors,
=find problems,
=assess the impact of errors,
=check error sources,
= visualize connections,
=and confirm the effects of improvements.
An error collection list (or data collection sheet) is either a piece of paper or part of a software that is used to?
document and display errors.
It is helpful to detect causes or patterns of problems. Before it can be used, typical errors should be ?
noted and clearly described.
The person who documents the error needs be able to do so without?
adding more errors.
Moreover, it is not very helpful to have a separate open category in which “other” errors are ————. This vague category does not help with future
—————————————-.
documented
improvement or prevention
If an “other” category is indeed added, it is better to ?
and ?
document the type of error to track its frequency
and decide whether it should be added to the typical error list.
The error list should also contain a time frame in which the error should be documented, e.g.,?
every hour, day, or week.
It should ideally contain information on:
the process,
component or faulty part,
date, time,
and person who documented the error.
Also helpful is a graph that shows?
the checked component and the typical errors that usually occur.
Overall, error collection lists are very easy to use and provide information about?
the problem to help find the cause.
table
The table above shows an example of an error collection list of ?
a Silgranit kitchen sink without a picture or graphic.
Silgranit is :
a durable, scratch-proof material containing a mixture of about four-fifths granite and one-fifth acrylic and ceramic (Blanco, n.d.).
The error list contains:
information on the product, process, place, and examiner.
The effort of using error collection lists is very low, because?
errors are simply counted and documented, and the benefit is great.
Statistics, especially histograms, can be compiled from ——————–very easily
the data
Flow chart
A flow chart is a diagram that describes and represents a process or system.
Business processes contain:
input,
activity or value-added,
and output.
A flow chart is
a graphical representation of these business process and shows how numerous different sub-processes interact with each other to complete the whole process.
Flow charts are commonly used in business because?
they are easy to use and very helpful to see the “big picture” of a business process. Some even call it “one of the best techniques of quality control” .
Besides providing the big picture, a flow chart can also?
A simple flow chart is laid out the following way.
a lot of detail. Flow charts can be used in quality control and in the planning process.
It contains an input (e.g., a document or someone placing an order of raw material) and always ends with an output (e.g., customer receives meal or a product is manufactured).
In between, there are:
-numerous activities that need to be performed, -decisions to be made,
-sub-processes to perform,
-and checks to undertake.
The input and output are:
processes displayed in an oval shape,
activities in a rectangle,
and decisions in a diamond.
The representation is standardized and explained in DIN 66001
table
The graph shows a hypothetical simple process.
It shows that the machine should only be started once it is on.
The start button should not be pressed when?
the machine is not turned on,
and therefore a decision loop is included to make sure that the machine is turned on.
Pressing the start button before the machine is turned on could harm the machine; thus,?
turning the machine on beforehand is part of quality control.
This is a simple example,
yet flow charts are very helpful for quality management, e.g., when new personnel are hired, they can get a good understanding of processes and tasks.
To create a flow chart, use the following steps
=Understand and identify all steps in the process. Order the steps chronologically.
= Lay out the steps in the flow chart.
=Review the chart with someone experienced with the process.
=In addition, we can also assign responsibilities for the processes in a flow chart if that is desired or needed.
There are many software programs that ————the representation of flow charts. A quick search in an internet browser provides several —————————————-.
facilitate
free solutions
Histogram
A histogram is a type of graph that displays rectangles of different lengths to show how frequently each value in a group of data occurs within different ranges.
A histogram is a ?
column diagram and a basic tool for displaying statistics.
It shows the frequencies of quantitative data.
Histograms are used to display continuous numerical data.
The goal is to ?
the data are distributed symmetrically or whether they are skewed to the left or right. The graph also shows the maximum and minimum value in the graph (Luthra et al., 2021, p. 41).
represent the distribution of the data and to understand the data properties.
After analyzing the histogram, we see whether?
the data are distributed symmetrically
or whether they are skewed to the left or right.
The graph also shows the maximum and minimum value in the graph.
graph
The data are divided into smaller sections (classes). The frequency of the data is then?
counted in a frequency distribution table or taken from the error collection list.
From this table, vertical bars are formed on the x-axis.
The frequency is shown on the y-axis, and the height is proportional to the frequency of occurrence in the data.
The histogram in the graph above was constructed with statistical software.
It shows the frequencies of 250 average process times (N=250).
The distribution has a mean process time of 162.29 minutes and a standard deviation of 46.57 minutes.
The y-axis on the left shows the frequencies: The higher the bar, the greater the frequency.
The x-axis shows the average duration of process times in designated classes.
There are several guidelines to follow when constructing a histogram (Foster, 2017, p. 273):
-The width of the histogram bars must be identical (e.g., 0—4, 5—9, 10—14).
-The classes must be mutually exclusive (e.g., 0—5 and 5—10 are not possible because 5 would be assigned to two histogram bars) and all-inclusive (e.g., 0—4 and 6—11 are not possible, because 5 is not included in the class representation).
-The minimum number of classes in a histogram is calculated with the following formula: \n\n\n\n\n\n\n\n \n \n \n \n\n where n represents number of raw data values and k represents the number of classes.
The number of classes can be retrieved by restructuring the formula to the equation:
picture
In the example above, 250 average process times were measured. By applying the formula, we receive at least eight classes (as shown in the graph)
picture
Generally, the number of classes should be above
————–. When using fewer than five classes, the form of distribution does not make much ———————————————.
five
sense or add value
The width of the classes is calculated by dividing the range (maximum value – minimum value) by the number of classes:
picture
The classes are then built by starting with the minimum value (here usually a slightly lower value is used to?
start the first class) and adding the class width: class 1 [53—88], class 2 [88.1—123.1], class 3 [123.2—158.2], class 4 [158.3—193.3], class 5 [193.4—228.4], class 6 [228.5—263.5], class 7 [263.6—298.6], and class 8 [298.7—233.7].
Statistical programs, like SPSS, or spreadsheet software, like Excel, help with the representation of histograms. Because?
Excel or comparable spreadsheet programs are installed on almost every computer and free alternatives exist (e.g., OpenOffice), there are virtually no costs to using histograms for quality control.
A Pareto chart (or dual axis graph) is used to?
analyze causes of problems, e.g., in the production process.
The Pareto principle is sometimes also called the
———————–, which means that ———- percent of the causes lead to ——– precent of the problems.
20\/80 rule
20
80
it is beneficial to sort problems or errors by their occurrence frequency.
Because of this rule,(The Pareto principle)
The few causes that lead to the majority of problems should ideally be?
improved or eliminated.
picture
The dual axis graph above refers to the errors in the production of kitchen sinks previously discussed in this unit. It shows the error types (missing part, color deviation, scratch, dent, and inaccurate drilling) on the x-axis. The y-axis on the left side refers to
the number of errors, represented by the blue bars, which are ordered by the number of errors in descending order from left to right. The y-axis on the right side refers to the cumulative percentage of errors, represented by the black line. The red dotted line is drawn for easier understanding of the Pareto chart.
It intercepts the y-axis line at 80 percent cumulative percentage. It seems as if most of the errors are due to missing parts and color deviations.
To be more precise, 80 percent of the errors are to the left of the dotted line, and 20 percent of the errors on the right of the dotted line.
Scratches, dents, and inaccurate drilling make up about 20 percent of the errors.
After looking at this Pareto chart, the firm should try to improve the processes, specifically to reduce the number of missing parts and to improve the number of kitchen sinks that deviate in color.
The Pareto chart is used widely because?
firms can expend little effort and receive a high value in using it.
They can set priorities for improvement and justify the order objectively
The graph can either be constructed with ————————————————–, so again, the tool can be easily applied by virtually ———————-.
statistical software or with Excel
every firm
A correlation is a statistical analysis that measures ?
the linear relationship between two variables. The correlation analysis helps explain the strengths and the direction of the relationship.
For example,
age and body height correlate positively with one another.
A correlation only gives information about the strengths and direction of the relationship between?
the variables and does not draw any conclusions about the causation (cause-and-effect).
With a correlation analysis, one cannot predict the influence of one variable on another. One needs a sufficiently large sample of at least ?
30 measurements.
Yet, larger samples of at least—————————— measurements are preferred.
A scatter plot can provide a —————————— of how the data are distributed and predict the relationship.
50 to 100
first indication
picture
If we require more than a simple prediction, it is possible to measure ?
the relationship between the two variables. Pearson’s correlation coefficient is used to calculate the linear relationship of two variables, statistically.
The Pearson’s correlation coefficient is the covariance of both variables divided by?
the product of their standard deviations.
the product of their standard deviations. For a sample, the formula is ?
picture
picture
The coefficient can take a value between?
—1 and +1. Values around 0 indicate that there is no correlation between the variables.
Values between +\/—0.1 and +\/—0.3 indicate a
———- correlation; values between +\/—0.3 and +\/—0.7 are —————— correlations; and high values in ————– terms (either close to +1 or —1) indicate a strong correlation between the two —————-.
weak
medium
absolute
variables
The correlation can easily be calculated with?
Excel, other common spreadsheet programs, as well as statistical software.
Correlation analysis is sometimes used to?
start the error elimination process, even if no cause-and-effect can be found by using this method.
Cause-and-effect diagrams are also called Ishikawa diagrams because?
they were developed by the Japanese quality expert Ishikawa.
The graph is sometimes referred to as?
the fishbone diagram, because of its shape.
The main idea is to?
find solutions to problems and portray these in a structured way.
The diagram thus displays suspected or proven causes of a problem and distinguishes between?
main causes and reasons.
To construct this type of diagram, :
-we first place the problem on the right side of a piece of paper.
-The second step is to either create categories or use Ishikawaʼs.
equipment, process, people, and raw material.
-Originally, Ishikawa used four main causes when applying the cause-and-effect diagram in the context of production:
equipment, process, people, and raw material.
Today, the environment and management are often added as?
potential categories.
An adequate template of the model is shown in the diagram below.
Next, a brainstorming session is?
conducted to identify the root causes of the problem and assign them to the categories.
One needs to assess each cause carefully and plot these in the diagram.
After the diagram is completed, it needs to be?
analyzed.
If the same cause appears several times, it is easy to see?
how to proceed with quality improvement.
In addition, a weight assessment can be used to?
prioritize problem diagnosis, where each person in the team can give points to solutions; the solution with the most points is selected.
The diagram is combined with other quality tools to either?
first diagnose the problem or find the causes.
There is no special software tool or expert knowledge required to?
construct cause-and-effect diagrams.
Quality control charts are among the oldest and most important tools to?
control quality.
These graphics display a series of statistical numbers (e.g.,?
upper and lower limits) taken from a sample
The goal is to visualize sample parameters, e.g., ?
mean values and data spread, and examine the behavior of a process.
With the help of the graph, it is easy to see if the process is under control. If the process is not under control, we can?
react quickly, find the cause, and eliminate the problem.
A sample may be taken, for example,
on a regular basis in the production process.
The x-axis represents the time when the sample is taken (e.g., 1, 2, 3, and 4).
The time frame depends on how often variations take place, e.g., hourly, daily, or weekly. The values from the sample are then plotted on the y-axis of the graph. If the deviation from the specified limits is too big, it is possible to correct or regulate the process.
picture
The center line (CL) is determined by?
previous data, e.g., a run in the test phase.
The upper (UCL) and lower control limits (LCL) are:
determined statistically by the accepted error rate and required specifications.
When a kitchen sink is produced and its size is beyond the upper specification limit, as seen at sample time 8 in the graph, it is considered ?
scrap and needs to be reworked or discarded. UCL and LCL are often the
mean value +\/—3 sigma (σ standard deviations) around the target mean.
Values below LCL and above UCL are typically seen as —————, which means that something is out ———————.
problematic
of control.
The upper (UWL) and lower warning limits (LWL) are often set to?
the mean value +\/—2 sigma (σ standard deviations) around the target mean.
The upper specification limit and the lower specification limit usually lie +\/—6 sigma (σ standard deviations) around?
the target mean and therefore represent nearly zero defects.
Quality control charts need to be?
examined after every entry, otherwise it is impossible to react to deviations quickly.
The main objective of using process control charts is to ?
“ensure the quality of the process,” especially when there is variation in the process, and thus tolerance limits are helpful.
If the warning line or control limit is exceeded in?
a sample or if an upward or downward trend is recognized, it is advised to sample more regularly to check whether the action to eliminate the problem was successful.
Ideally, the quality control chart also includes:
flow charts with decision rules, so that common problems can be eliminated quickly
Although quality control charts are helpful to spot problems, they are?
a reactive tool
The problems can only be eliminated after?
limits have been breached and the problems have already occurred.
Only if trends are visible, can one foresee future problems and eliminate them before?
they occur.
Managers have a variety of tools they can use to improve?
processes and products.
This section explains several management tools that help?
support a firm to solve problems.
first introduced these primarily visual tools (also called M7) in Japan in 1978 was?
The Japanese Union of Scientists and Engineers (JUSE).
These tools are also called?
the “new seven tools for quality control” or “new seven”.
They are usually applied during group work in the planning and development phase of a project when?
numerical data are not yet available.
In addition, the tools support the quality control tools that were discussed in ?
the previous section.
Brainstorming
The process of finding the (best) solution to a problem by collecting spontaneous ideas is commonly known as brainstorming.
When teams use the brainstorming technique to develop ideas, these are initially rather?
unstructured, and there is need to put them in order.
An affinity diagram helps :
structure ideas, see relationships, and develop new approaches toward solutions.
table
According to Foster and Benes and Groh, there are a series of steps to create an affinity diagram in a team environment after material has been collected in a brainstorming session:
1-Identify and spell out the problem in a short and clear statement that everyone in the group understands.
2-Give group members about ten minutes to write issues related to the problem on note cards. Make sure only one idea is written on each card.
3-Lay out note cards on a flat surface and let team members quickly group the cards into similar themes or clusters silently (without discussion).
Continue to move cards silently into piles until consensus is reached.
4-Discuss headings of the sub-issues thoroughly and create header cards.
5-These headers then create options for the problem solution.
6-Draw an affinity diagram of the solution and order the solutions according to their importance using a color-coded sticker, e.g., red (extremely important), yellow (important), and green (less important).
7-Provide each team member with a copy of the diagram, e.g., take a picture of the original and attach it to the protocol.
An affinity diagram can be constructed in a team session and improve team quality. No extra software is necessary for its development. It is a very quick and easy way to find ideas in a communicative way.
The interrelationship diagraph helps?
identify solutions to problems.
Like the affinity diagram, it is another graphical representation to :
combine facts,
relationships,
and arguments that cannot be displayed with numerical numbers.
In particular, the interrelationship diagraph allows the connection of?
dependencies and their weighting as driving forces or outcomes.
To construct an interrelationship diagraph, follow these steps (Benes & Groh, 2017, p. 281; Foster, 1996, 2017, p. 286):
1-Phrase the problem in a clear sentence and place it in the middle of a bulletin board. The prior construction of an affinity diagram helps to identify all the issues that relate to the problem. Put these issues on sticky notes or note cards and add these to the bulletin board.
2-Examine the issues and create the graph. Use one note card at a time and check if there is a relationship between issues on the bulletin board, e.g., ask “what other issues on this diagraph are caused or influenced by this issues?” (Foster, 2017, p. 286). Draw arrows for the identified causes and effects. Never use double-pointed arrows. Continue working until all issues are addressed. It is helpful to first draw a cause-and-effect (fishbone) diagram to be clear on the relationships (Herrmann & Fritz, 2018, p. 163).
3-Quantify the issues by counting how many arrows point toward the note. The card with the most arrows pointed toward it is a “key factor” (Foster, 2017, p. 286). Evaluate the key factors. Note that performance indicators are those with many incoming paths, and root causes are those with many outgoing paths (Foster, 2017, p. 286).
