Interface for a Shopping Information Device
Topic:
We have decided to design an interface for a portable
information device that can be used to present relevant information
about store items to customers and store clerks in a large scale
department store such as Home Depot. This module will provide
information on the location, the availability, and description of
the item the customer is looking for.
Problem:
Currently a customer unfamiliar with the location of the items
in a store is forced to spend most of his or her time in searching
for the item or getting employee assistance. This may not be a
concern in a small store but in many of today's large scale
department stores searching for an item without any prior knowledge
of its location or finding help from an employee can be both time
consuming and frustrating - in practice this the customer can spend
much time on walking through aisles after aisles of goods before
finding the item wanted or employee who can help.
The proposed interface addresses both these problems directly
by providing information on the items in the store based on the
customer's input such as item's description and item's category,
thereby saving customer's time as well as the store employee's
time.
The challenges of designing an interface for this device
includes:
- usability: since the user base can be as broadly defined as
the customer of a large department or warehouse store.
- restriction of interface space: since the device is
portable, its interface must provide enough information in a
limited space.
Requirement Specification
Needs Analysis:
Customers who are in a store often spend much time on
familiarizing themselves with the store layout. This can lengthen
the time it takes to find the items they are looking for. Also, in
a large store, looking for an item can take much more effort and
time than the customer would like to spend. While it is possible
to find help from employees, it can also be time consuming if the
demands for help is greater than the supply, i.e., the store might
not have enough employees to handle the customers quickly.
The proposed system uses a portable device to provide the
customers with information such as the location, availability, cost
and use of the items they are looking for. This can decrease the
time the customer spends on searching for the item and/or waiting
for help from the store employees. For example, if a customer wants
to know the price of an item, he or she does not have to go to the
location of the item to find it, instead, it is presented to the
customer on the spot. In the current system, to find the price of
an item, a customer either has to go to the location of the item
or ask a store employee.
User Analysis:
Users of the proposed interface are the customers who shop
at large department or warehouse stores and the employees of those
stores. To collect information about the users, the techniques of
personal interviews and observation were used. One of the branches
of a large store (Home Depot) was selected as a place to conduct
personal interviews and observation.
The data collected from the techniques mentioned above point to
a user base comprising of both males and females, both in the age
group of about 15 to 70 years. They could be classified broadly
into novice, intermediate and expert users. An expert user knew
exactly what he wanted and went to the location of the item without
any assistance from the employees of the store. An intermediate
customer had an idea of what he was looking for but was unable to
locate it. He had to get the help of a store employee to locate the
item. A novice customer did not have an exact description of the
item he or she was looking for but had an idea of its usage. He had
to get a lot of help from an employee before getting the item
needed.
Most of the customers observed belonged to the intermediate and
novice categories and had varying levels of difficulty in locating
the items they were looking for and some of them wanted an alternate
method of locating the items. Interviews with store employees
confirmed these factors.
Task Analysis:
To find an item in a store on his/her own, a customer must
utilize information provided by the store, i.e. signs describing
the item category for an aisle or the section of the store he or
she is in. Whenever the customer is not provided or cannot find
the information needed, he or she is required to seek help from an
employee. We have decomposed the task of locating an item in a
store and created a listing of the sequence of tasks in List 1. We
have also created a hierarchical description of this list and its
plans in Fig. 1.
Functional Requirement:
Based on our analysis, we have decided that the users of our
interface will be able to specify the item they are looking for by
providing any of the following:
- name of item
- category of item
- key words relevant to item's description and use
- select from choices given by the interface
Since our user analysis indicates that the majority of
customers do not know exactly where an item is located but have
either the name of the item they are looking for or a description
of a mental picture of the item they want, they should be able to
search for the item by its name or a key word in its description.
By the same token, the user may have an inkling of the category the
item may be classified as and they will be given possible choices
of items in the categories. For example, if a user needs shower
curtain rings but cannot remember what they are called, he or she
can try to located it by browsing through items related to bathroom
perhaps, or items with the key word rings in their description.
Finally the user will be able to select from the possible items the
one he or she actually need and get information on the item.
Non-Functional Requirement:
Our interface will reduce the time users spend on seeking help
from store employee and searching for items by determining the
information needed by the user and provide the information to him
or her. The interface must be portable so the user can have the
information while shopping in the store. The information provided
will also be updated in real time so that when the items are
exhausted by other users, its availability can be modified and
provided to the user. The prices and locations of the items can
also be updated accordingly.
List 1:
0. go to the store
1. look for the item (without help from store employees)
1.1 decide category of item
1.1.1 use knowledge of item to decide on category
1.1.2 if unable to decide category then ask for help
1.2 locating the aisle containing the category
1.2.1 follow map if there is one, else walk by each aisle
and look at the aisle description
1.2.2 if unable to locate aisle, ask for help
1.3 look for the item in that aisle
1.3.1 look at each item or group of items and locate item
needed
1.3.2 if unable to locate item, ask for help
1.4 if not found repeat a, b, and c on another possible
category
1.5 go to step 3
2. asking for help
2.1 get assistance from store employees
2.1.1 look for employees in the immediate area
2.1.2 ask for information on locating item or more help
2.1.2.1 if employee cannot provide information on
item, get reference to other employees
and repeat step 2.1.2
2.2 if employee knows item is not there then go to step 4
2.3 go to aisle
2.4 look for item
2.4.1 if have the location of item within the aisle, go
to the location, else go to 1.3.1
3. get item (done): result have item
4. (done): result no item
The Design Techniques Used
In creating the user interface requirement specification, we
relied heavily on the task analysis of the current system of
obtaining information of an item in a store. In this part of our
requirement analysis we used a hierarchical decomposition of the
tasks involved as well as a plan based list of tasks to be performed
by a customer when he or she tries to find out information on a
store item. We did not choose to use a knowledge based or the
entity-relationship analysis at this point since the task at hand
is procedural oriented.
From the result of the task analysis we then made decisions
on what type of information to provide the customers to reduce the
time they spend in locating and finding information on the items
they are looking for. Since we observed from the task list that
whenever a customer cannot find additional information from the
environment, he or she will be forced to seek help from store
employees.
We have also relied on interviews with customers in Home Depot
to give us an indication of the possible user profile and needs.
Due to the small size of our group, we have decided that a
questionnaire approach will be the most effective way to profile
the users. From the user analysis it appears that all except the
most expert users, who know exactly where every item in the store
is located, their prices, etc., will need some form of help from
an employee (who can also be considered an expert). It is clear
that much time may be saved if this part of the task can be
streamlined, i.e. the information needed by the customer can be
provided by our interface instead.
ABOUT OUR GROUP
Our group consists of five students with different backgrounds and
skills:
Raj Chappidi
A graduate student in electrical engineering, Raj concentrates
his studies in the areas of computers and communications and stud-
ies courses in computer science at a minor level. His programming
experience includes use of Assembly, C, Fortran and Matlab. He is
currently working as a programmer at American Megatrends Inc. He
uses Intel assembly language to work in the area of BIOS.
Dave Crater
Dave is also a graduate student in electrical engineering. He
received his B-EE from Georgia Tech in the spring of 1994 and is now
concentrating on computer engineering, digital signal processing,
and electronic design. His programming experience includes
Assembly, C, Fortran, and Pascal. Dave has worked as a computer
programmer with the Department of Defense for the last three years,
helping to automate their extensive personnel databases and create
enhanced user interfaces. He is now a commissioned officer in the
U.S. Air Force, and is scheduled to begin active duty in Colorado
Springs in July 1995.
Whitney Foster
Whitney Foster is a Masters student in the College of Computing. His
undergraduate degree is in Applied Mathematics with a concentration
in Computer Science from Carnegie Mellon. He came to Atlanta in
1989 to work for Lockheed on the Pilot's Associate program (a user
interface to link the pilot of a single seat aircraft to his vehicle
and to help improve his performance and survivablility). He is
currently working on a variety of other programs with an emphasis on
simulation.
Will Luo
A graduate student pursuing a master's degree in the college of
computing, Will has a background in mathematics and physics. He
has programming experience in C, Pascal and Assembly languages.
He is also a co-op with IBM in their multimedia development group.
M.Narayanaswamy (swamy)
A graduate student pursuing a master's degree in the college of
computing, Swamy has a background in electrical engineering. He
has programming experience in C, Assemble, Prolog, Lisp, Small-
Talk, Cobol and Sql languages. He has industrial experience in the
area of software development.
Combined skills:
The team has combined skills in hardware and software. The
knowledge gained in this course will enable them to design human-
computer interfaces in a creative and efficient manner.