Scribd
Upload
56 views27 pages

S08Proj2 IntroToProject

This document discusses queue theory and its application to managing ATM lines. It provides examples of queues that occur in different situations. The document then outlines a class project for a bank to analyze queue length and wait times at their ATMs to determine the optimal number of ATMs and queueing model to meet various customer service goals. Objectives of the project include validating advertising claims about wait times and queue length based on historical ATM usage data.

Uploaded by

Pranshu Agrawal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
56 views27 pages

S08Proj2 IntroToProject

This document discusses queue theory and its application to managing ATM lines. It provides examples of queues that occur in different situations. The document then outlines a class project for a bank to analyze queue length and wait times at their ATMs to determine the optimal number of ATMs and queueing model to meet various customer service goals. Objectives of the project include validating advertising claims about wait times and queue length based on historical ATM usage data.

Uploaded by

Pranshu Agrawal
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 27

Project 2: ATM’s & Queues

ATM’s & Queues


 Certain business situations require customers
to wait in line for a service

 Examples:
 Waiting to use an ATM machine
 Paying for groceries at the supermarket

 A line of people or objects is called a “queue”


ATM’s & Queues
 Queues occur in many places:

 Running multiple programs on a computer


 A print queue is formed when many documents are
sent to the printer
 Telephone calls on a switchboard
 Vehicles waiting at a traffic light
ATM’s & Queues
 Studying how these lines form and how to manage
them is called Queuing Theory

 Queuing Theory has become an important tool in


business decisions regarding quality and expense of
customer service

 Example: Supermarket manager sees checkout lines are too


long, so more cashiers are called to work the registers, but
this costs more money
ATM’s & Queues
 Automated services make queue theory
important when direct monitoring of service
isn’t possible
 Example:
 Bank manager can’t monitor ATM machine service at
mid-night.
 Opening up more machines might improve customer
service but may cost a lot of money
ATM’s & Queues
 Managing queues is a balancing act:

Customer $$$$$$$$$
Satisfaction $$$$$$$$$
ATM’s & Queues
 Two Queue Models

 Standard Queue

 Serpentine Queue
ATM’s & Queues
 Standard Queue

 Customers select what they believe to be the shortest


or most rapidly moving line from individual queues at
several stations.
 This model is used at most supermarkets.
ATM’s & Queues
 Serpentine Model

 Customers form a single


line, and advance to the
front to get their
service.
 Used at most airline
ticket counters and in
many post offices
ATM’s & Queues
 Analyzing how to manage queues often uses
computer simulation

 Two types of Simulation

 Monte Carlo

 Bootstrapping
ATM’s & Queues
 Monte Carlo Simulation

 Sample data is used to estimate the actual


probability distribution of some random variable.

 This theoretical distribution is then used to


generate new samples.
ATM’s & Queuing
 Bootstrapping

 When the data does not indicate any known


theoretical probability distribution, we can
simulate new data by random sampling from the
original data
ATM’s & Queues
 Class Project

 The People’s Bank has 3 ATM’s


 At least one ATM is available 24 hours a day 7
days a week
 Bank manager has records of ATM usage and
customer service times for 5 weeks
ATM’s & Queues
 Mean numbers of customers arriving for ATM
usage during every hour of the week is
contained in Queue Data.xls.
 The complete arrival data for the 9:00 a.m. and
9:00 p.m. hours on Fridays are shown in that
file as well.
 These hours happen to be the bank’s busiest
days of service.
ATM’s & Queues
 We will study the queues for the ATM’s during:
 The 9:00am hour on Friday
 The 9:00pm hour on Friday

 The starting and ending times of ATM service were


recorded for each arriving customer.
 Data for these service times during the first week of
record keeping are shown in Queue Data.xls.
ATM’s & Queues
 Bank manager wants to avoid long wait times, long
queue lengths, and do this using the least number of
ATM’s

 The bank manager would like to know what level of


service to provide for managing the queues based on:

 Services Times for individual customers


 The number of customers waiting to be served
ATM’s & Queues
 Terms:
 Wait Time (in min): The period of time that a customer must
wait between arrival and the start of his or her access to an
ATM
 Delayed: A person who must wait more than 5 minutes
 Number in Queue: the number of people in line waiting
before an arriving customer can reach an ATM
 Irritated: queue length is more than 3 customers
 Total Present: the total number of patrons present in the
queue
ATM’s & Queues
 The bank manager is looking at three
advertising claims for service times:
 (Mean Wait Claim) The mean waiting time is at
most 1 minute.
 (Maximum Wait Claim) No one will wait more than
12 minutes.
 (Percent Delayed Claim) At most 5% of the
customers will be delayed (wait more than 5
minutes)
ATM’s & Queues
 The bank manager is also looking at three advertising
claims for the number of customers waiting in line:
 (Mean Queue Claim) The mean number of people in the
queue will not exceed 8.
 (Percent Irritated Claim) At most 2% of the customers will be
irritated (find more than 3 people in line or waiting to be
served).
 (Maximum Present Claim) The total number present will
never exceed 10.
ATM’s & Queues
 Project Assumptions:
 No one is using an ATM or waiting for a machine at
the start of the hour.

 Service times for each ATM have the same


distribution as sampled in Week 1 Service Times in
the sheet Data of Queue Data.xls.
ATM’s & Queues
 Project Assumptions (cont)
 The time until the first arrival and the times
between arrivals of customers have the same
distribution.
 In the standard queuing model, if more than one
ATM is open, arriving customers enter the shortest
of the existing queues. If two or more queues are
the same length, a customer selects a queue at
random.
ATM’s & Queues
 Objectives:

 Based only on 9 a.m. hour on Fridays, how many ATM’s should be


opened and what queuing model should be used to validate each
advertising claim during 9-10 a.m. period?
 Based only on 9 p.m. hour on Fridays, how many ATM’s should be
opened and what queuing model should be used to validate each
advertising claim during 9-10 p.m. period?

NOTE: We only consider the use of a serpentine model when three


ATM’s are in use
ATM’s & Queues
 Objectives (cont)

 Finding the hourly cost of a gift certificate program for 3


ATM’s Serpentine:

 If a serpentine queue is used, customers don’t physically


stand in a line because the bank currently uses a number
dispenser and service indicator that gives customers slips of
paper indicating their position in the queue.
ATM’s & Queues
 Objectives (cont)

 Finding the cost of gift certificate program (cont)

 Bank is considering updating to a system that stamps the


arrival time of a customer which could be used to document
a customer’s wait time

 The hourly cost for such an upgrade (maintenance,


purchase price, etc.) is $20
ATM’s & Queues
 Objectives

 Finding the cost of the gift certificate program (cont)

 A $25 gift to any customer who is delayed (waits more than


5 minutes).

 What is the expected hourly cost of such a plan? How


would this change if it is estimated that only 60% of eligible
customers would decide to claim the gift?
ATM’s & Queues
 Team Data will be posted on class web page

 Data includes

 historical records of ATM service times and customer arrival times


for two hours out of each week

 Parameters for six potential advertising strategies.


 Mean Wait Claim
 Maximum Wait Claim
 Percent Delayed Claim
 Mean Queue Claim
 Percent Irritated Claim
 Maximum Present Claim
ATM’s & Queues
Team Data will be posted after Spring Break

Team Preliminary Report


 Date: Monday March 31st, 2008

You might also like