Environmental Engineering Reference
In-Depth Information
Process
Operators
Primary
ingredient
Intentional
equipment
damage
Unmotivated
workers
Moving the
product
Washing and
rinsing
Lack of
training
Lack of
cooperation
Leaking
equipment
High water
use
Lack of
maintenance
Excessive
dirt in raw
material
Reluctance
to invest in
repairs
Lack of
leadership
Materials
Management
Figure 5.8
Example of cause-and-effect diagram (or fishbone diagram) to study the sources of water
consumption in a food plant.
excessive water use (the problem) in a fruits and vegetable processing facility. This
particular example has identified four main factors and several causes for each individual
factor.
Histograms are frequency distribution diagrams used for the representation of numerical
data. Data of a particular variable is divided in nonoverlapping intervals and the frequency of
occurrence represented as a bar in a two-dimensional diagram.
A control chart is a graphic representation of a variable in time. The main objective of
control charts is to monitor whether a process variation is within the expected limits (process
in statistical control) or the variation is the result of special causes.
Pareto diagrams, which have been defined in a previous section, are powerful in identifying
the “vital few [problems] from the trivial many” as it was described by Dr. Joseph Juran. The
“vital few” follow frequently the 80/20 rule, which states that 80 percent of the problems can
be attributed to 20 percent of the causes.
Check sheets are simple data collection systems for numeric or categorical data at the site
where the data is produced. Check sheets are forms specially designed to record the data by
making checkmarks or other symbols, so the finished graphical representation emerges with-
out the need of further analysis or data transcription. Table 5.3 presents an example of prob-
lems associated with water consumption recorded over the span of several weeks in a fruits
and vegetables processing plant.
Scatter plot is another name for
x/y
diagrams that depict graphically paired sets of data
from two numerical variables on perpendicular axes. The main use of this kind of diagrams is
the detection of a potential correlation between the two variables.
Stratification is a procedure used to cluster data into homogeneous groups and used in
combination with other tools. For instance, points in a scatter plot can be grouped by day of
the week, time of the day, equipment, shift, supplier, etc. Once the data is separated in groups
then the existence of patterns can be identified.
