From “Just-In-Time” delivery and inventory control in a warehouse, to customer service at a call center, to line management at Disneyland, sometimes knowing how many customers you have is not as important to know how fast they are piling up. In this Project you will need to collect real world data, use it to build a model, and then use that model to assess a second set of data that you will also collect.

First you need to decide what you are going to measure. It needs to be something that can be counted during a time interval. Examples include but are not limited to: Customers queuing up at a SUBWAYtm ; Cars entering a parking lot; Students walking past your dorm room window. You will take this count for an interval of at least one hour. For the first data set, you should have a count of at least five. (So you will need to use Personal Probability to make an appropriate choice of measurement.) Use this first count to construct a Poisson Distribution that models the number of “Events” per hour.

On a different day, make the same measurements, again for at least one hour.

• The count of the amount of customers going through the ASU hotdog stand in an hour:

  • DAY ONE: Tues: 02/18/2020

• 
  

  • 11:47- 1

  • 11:48- 1

  • 11:49- 1

  • 11:50- 1

  • 11:54- 1

  • 11:59- 1

• 
  

  • 12:00- 1

  • 12:02- 1

  • 12:03- 1

  • 12:12- 1

  • 12:34- 1

  • 12:35- 1

  • 12:47- 0

  • TOTAL- 12 people

• 
  

  • DAY TWO: Tues: 02/25/2020

• 
  

  • 11:47- 1

  • 11:49- 1

  • 11:50- 2

  • 11:51- 2

  • 11:54- 1

  • 11:56- 1

  • 11:57- 1

• 
  

  • 12:03- 1

  • 12:05- 1

  • 12:06- 1

  • 12:07- 1

  • 12:11- 1

  • 12:12- 1

  • 12:13- 1

  • 12:14- 1

  • 12:15- 1

  • 12:21- 1

  • 12:24- 1

  • 12:30- 1

  • 12:42- 1

  • 12:43- 1

  • 12:47- 0

  • TOTAL- 23 people

• Poisson Distribution (using the first count):

• If this new count is less than the average value given by your model, what is the total probability that your model would produce that value or less

• If it is greater than the average, what is the total probability your model would produce that value or greater?

• How does this new information affect your confidence in the original model?

• What additional factors might explain the similarities/differences between the two counts?