Basic HTML Version
© 2014 Pacific Crest
139
Step
Explanation
Watch it Work!
5.
Is it a well-known
distribution?
Does a commonly used
distribution exist or do we create
one? Check your graph against
off-the-shelf distributions.
Matches the Poisson Distribution
6.
Validate
Use additional sets of data,
preferably from a different context
exhibiting the same phenomenon,
to validate the distribution.
We check with data from a similar
interstate in another city and find the
model works well.
Scenario:
A manager wants to assess the efficiency of his staff and in particular know how
long it takes to complete a certain task.
Step
Watch it Work!
1.
What is the random
phenomenon?
The time by each employee to file a TPS report is a random
phenomenon and can be modeled with a probability distribution
2.
Determine the scope The phenomenon is observed with many tasks at most offices, but the
manager only has access to his staff for collecting data
3.
Get reliable data
The manager collects the following data
Time (min)
Frequency Time (min)
Frequency
0-14.9
1
60-74.9
2
15-29.9
10
75-89.9
3
30-44.9
57
90-104.9
7
45-59.9
13
105-119.5
1
4.
Construct a
Cumulative
Distribution
time (min)
frequency
0
10
20
30
40
50
60
Time (min) Relative Frequency
0-14.9
0.0106
15-29.9
0.1064
30-44.9
0.6064
45-59.9
0.1383
60-74.9
0.0213
75-89.9
0.0319
90-104.9
0.0745
105-119.5
0.0106
3.3 Probability Distributions