The Exponential Distribution - PowerPoint PPT Presentation

1 / 10
About This Presentation
Title:

The Exponential Distribution

Description:

A bank manager would like to use an analytical queuing model to study the ... Its highest bar is not the one farthest to the left but instead corresponds to ... – PowerPoint PPT presentation

Number of Views:32
Avg rating:3.0/5.0
Slides: 11
Provided by: lisa284
Category:

less

Transcript and Presenter's Notes

Title: The Exponential Distribution


1
Example 14.1
  • The Exponential Distribution

2
Background Information
  • A bank manager would like to use an analytical
    queuing model to study the congestion at the
    banks automatic teller machines (ATMs).
  • A simple model of this system requires that the
    interarrival times and service times are
    exponentially distributed.
  • During a period of time when business is fairly
    steady, several employees use stopwatches to
    gather data on interarrival times and service
    times.

3
Background Information -- continued
  • These are listed in the table on the next slide.
  • The bank manager wants to know, based on these
    data, whether it is reasonable to assume
    exponentially distributed interarrival times and
    service times.
  • In each case he also wants to know the
    appropriate value of ?.

4
(No Transcript)
5
EXPONFIT.XLS
  • To see whether these times are consistent with
    the exponential distribution, we plot histograms
    of the interarrival times and the service times.
  • The histograms appear on the next slides.
  • This file contains the model.

6
(No Transcript)
7
(No Transcript)
8
Solution
  • The histogram of the interarrival times appears
    to be quite consistent with the exponential
    density shown.
  • Its highest bar is at the left, and the remaining
    bars fall off gradually from left to right.
  • On the other hand, the histogram of the service
    times is not shaped like the exponential density.
  • Its highest bar is not the one farthest to the
    left but instead corresponds to the category 15
    to 30 seconds.

9
Solution -- continued
  • There is some minimum time required to process
    any customer, regardless of the task, so that the
    most likely times are not close to 0.
  • Therefore, the exponential assumption for
    interarrival times is reasonable, but it is
    questionable for service times.
  • In either case, if the manager decides to accept
    the exponential assumption, the parameter ? is
    the rate of arrivals and is estimated by the
    reciprocal of the average of the observed times.

10
Solution -- continued
  • For interarrival times, this estimate of ? is the
    reciprocal of the average in cell C39 1/25.3
    0.0395 that is, 1 arrival every 25.3 seconds.
  • For service times, the estimated ? is the
    reciprocal of the average in cell G39 1/22.3
    0.0448 that is, 1 service every 22.3 seconds.
Write a Comment
User Comments (0)
About PowerShow.com