Toggle Clamps
Toggle Clamps
Toggle Clamps
C. SEC Filings
Securities and Exchange Commission (SEC) files are available via
Bloomberg in EDGAR files, or some companies have access to their
SEC filings on their Web sites. In accordance with the Securities and
Exchange Commission Act of 1934, all companies that have publicly
issued securities must file various statements with the SEC and also
distribute them to the stockholders.2.2 Risk Categories
A company’s general risk management principles should be based on
identifying key risks, such as:
1. Strategic risks. Risks must be taken into account in strategies.
There is a correlation among strategy planning, decision making,
and risk analysis.
2. Operational risks. Sourcing belongs to operational risks. The quality,
correct timing, and scheduling of materials; optimal pricing;
and supply interruption can be categorized under sourcing risks.
A poor supplier base is a potential factor in sourcing risks.
3. Hazard risks. Hazard risks cause damage to assets (for example,
buildings or machinery) or intellectual assets (for example, brand
or product liabilities). Typically, hazard risks can be insured.
4. Financial risks. Financial risk is the threat of losing an asset or income.
Thus, financial risk involves two elements: (1) the individualThe range is often plotted on control charts
as discussed in Section C of this
chapter.
One of the disadvantages of using the range as a measure of dispersion is that
it uses only two of the values from the data set: the largest and the smallest. If
the data set is large, the range does not make use of much of the information contained
in the data. For this and other reasons, the standard deviation is frequently
used to measure dispersion. The value of the standard deviation may be approximated
using numbers from the control chart. This method is explained in Part 4,
“Process Capability,” of Section C of this chapter. The standard deviation can also
be found by entering the values of the data set into a calculator that has a standard
deviation key. See the calculator manual for appropriate steps.
Although the standard deviation is seldom calculated by hand, the following
discussion provides some insight into its meaning. Suppose it is necessary
to estimate the standard deviation of a very large data set. One approach would
be to randomly select a sample from that set. Suppose the randomly selected
sample consists of the values 2, 7, 9, and 2. Naturally, it would be better to use a
larger sample, but this will illustrate the steps involved. It is customary to refer to
the sample values as “x-values” and list them in a column headed by the letter
x. The first step, as illustrated in Figure 2.7a, is to calculate the sum of that column
x, and the mean of the column x–. Recall that
x
x
n
Median. The median is the value that has approximately 50 percent of the values
above it and 50 percent below. To find the median, first sort the data in ascending
order. If there is an odd number of values, the median is the middle value of
the sorted list. If there is an even number of values, the median is the mean of the
two middle values. The common symbol for median is x~, pronounced “x wiggle.”
Example: The list in the previous example is first sorted into ascending
order: 1 2 2 3 3 3 4 4 6 7 7 8 9
Since there are 13 values, the median is the seventh value in the sorted list, in this
case
x~ = 4
Example: Find the median of the six-element set 12.7 12.9 13.5 15.0 15.0 17.2
The values that represent the center of a data set are called, rather awkwardly,
measures
of central tendency. There are three commonly used measures of central
tendency: mean, median, and mode.
Mean. Mean is statistical jargon for the more common word “average.” It is calculated
by finding the total of the values in the data set and dividing by the number
of values. The symbol used for “total” is the Greek capital sigma, . The values of
the data set are symbolized by x’s and the number of values is usually referred
to as n. The symbol for mean is an x with a bar above it (x–). This symbol is pronounced
“x bar.” The formula for mean is
x
x
n
The Binomial Distribution. The prefix “bi-” implies the number two, as in bicycle
(two wheels) and bipartisan (two political parties). The binomial distribution is
used when every object fits in one of two categories. The most frequent application
in the quality field is when every part is classified as either good or defective,
such as in valve leak tests or circuit continuity tests. In these cases there are two
possibilities: the valve either leaks or it doesn’t; the circuit either passes current or
it doesn’t. If data on the amount of leakage or the amount of resistance were collected,
the binomial distribution would not be appropriate. The word defective is
often used in binomial distribution applications. A typical example might consider
the number of defectives in a random sample of size 10. Notice that the number
of defectives is not a continuous variable because, for instance, between two
defectives and three defectives there are not an infinite number of other values,
that is, there can’t be 2.3 defectives. That is why this distribution is called discre
Control Charts
One of the disadvantages of many of the tools discussed up to this point is that
they have no time reference. The chart may clearly demonstrate that a process had
a problem but gives no clue as to when the problem occurred. This section will
illustrate a number of time-related techniques. The discussion will culminate with
control charts.
The Run Chart. If a specific measurement is collected on a regular basis, say every
15 minutes, and plotted on a time scale, the resultant diagram is called a run chart.
An example of a run chart is shown in Figure 1.13.
One of the problems of the run chart is that the natural variation in the process
and in the measurement system tends to cause the graph to go up and down
when no real change is occurring. One way to smooth out some of this “noise” in
the process is to take readings from several consecutive parts and plot the average
of the readings. The result is called an averages chart. An example is shown in
Figure 1.14.
One of the dangers of the averages chart is that it can make the process look
better than it really is. For example, note that the average of the five 3:00 p.m. readings
is .790, which is well within the tolerance of .780–.795, even though every
one of the five readings is outside the tolerance. Therefore, tolerance limits should
never be drawn on an averages chart. To help alert the chart user that the readings
are widely dispersed, the averages chart usually has a range chart included
A Pareto chart of this data would list the causes on the horizontal axis and the percent
of occurrences on the vertical axis. The causes are listed in order of decreasing
number of occurrences. The Pareto chart is shown in Figure 1.11.
The Pareto chart shows that the people working to reduce the number of
occurrences should put their main efforts into preventing outages caused by animals.
If they expended a lot of resources preventing transformer leakage and were
able to eliminate it completely as a cause of power outage, it would still solve only
about six percent of the occurrences.
A Pareto chart often shows one source as the overwhelming cause of defects.
However, in some cases it may be necessary to do some creative grouping to obtain
a single cause that accounts for the bulk of the problem. Suppose a team is seeking
to reduce the number of defective valve stems from a multi-stage machine operation.
They gather the following data on the types of defects observed:
conflicts (see Figure 4.9). In the end, however, all parties must
consent to the WBS before building a schedule - a task that
can also try anyone's patience.
Building a WBS also threatens people (see Figure 4.10).
They do not want to list the work that they will perform. They
may fear being held accountable or constantly being tracked.
In other words, they feel that they may lose their autonomy.
Many team members, especially those who are prima donnas,
feel that way. The client often wants a good WBS so it can
track progress, that is, as long as it is not being tracked. Senior
management will express an interest in a WBS but only at a
high level. The project manager prefers a WBS to generate