October 16, 2020

Dr. Jenahvive Morgan

College of Engineering
Michigan State University
219 Wilson Road
East Lansing, MI 48825

Dear Dr. Morgan,

Attached is a copy of Section 26, Team 5’s Project #1 Final Report for EGR 100, Fall 2020.

This report describes the design process and how the smartphone application functions. The
application helps people choose where to eat out nearby. The design of the app is described,
along with data that was collected through user surveys. This data was analyzed and its results
were listed in this report.

Using the MIT App Inventor, the brightly colored app has a “randomize” button that lists the
current and previously randomized restaurant. The restaurant preference selected along with
Google Maps is utilized to find the restaurant chosen by the app. Zomato is a website that
provides more information on the restaurant selected. This information includes the price and the
menu of the selected restaurant.

I hope this report meets your requirements, and any questions will gladly be addressed by email.

Sincerely,

Peter Gojcaj, Kenny Yue, Suhaimi Azam, and Jack Taylor

1
INTRODUCTION

No matter where anyone goes in the United States, they will likely find a form of fast food or
dine-in establishment not too far from where people live. ​Around 80% of adults in Michigan
visited fast-food restaurants at least once a month with 28% of them eating out more than or at
least 2 times per week [1]. In relation to food, cell phone apps connectivity to the internet is
important in ordering online as 20% of consumers preferred using online methods of ordering
food and 42% of consumers preferred a restaurant over others if it offered ordering online [2].
Cell phones have existed for decades, but the commonly known smartphone goes back to 1993
with IBM and it had its own cell phone applications such as a calendar and world clock. The use
of cell phone applications and ​many ​different restaurants creates confusion. The confusion leads
to people remaining hungry for longer and more frustration. Along similar lines, would-be
consumers who frequently visit a single restaurant and want to try something new may be stuck
on which new restaurant to try out. This is where the app addresses all of these problems. The
app partially takes the decision out of the hands of the consumer and allows a random restaurant
in the East Lansing area to be selected through different preferences or no particular preference.
This addresses the problem of not knowing where to go as it gives the user a random choice
along with the previous random choice each time the location is randomized for comparison. It
also allows people to find new preferences in where to eat or try a new restaurant they never
went to under the same preference to a restaurant they go to regularly. Finally, the app allows a
website, Zomato, to be pulled up to find reviews, addresses, and menus on each restaurant.
Analyzed data proves that the app succeeded in solving the given problems.

METHODS

The app was built in MIT App Inventor and utilizes App Inventor User Interface components
such as the spinner, the button, and the label. This allows the user to interact with the app as seen
in Figure 1 below.

Figure 1​: Picture of the app

2
The user is able to select the restaurant preference by using the spinner component, as seen in
Figure 2. The user selects from five preferences: Fast Food, Expensive Dine In, Delivery, and No
Preference, which allows the app to randomly choose restaurants based on preferences.

Figure 2​: Preference on the spinner

When the Randomize button is clicked it selects a restaurant randomly from one of the three
lists, as seen in Figure 2, and places the restaurant in the selected place label. If there is currently
a restaurant in the Selected Location label the restaurant is moved to the Previous Location label,
as seen in Figure 3.

Figure 3​: Selected location to the previous location

3
6 global variables were used, as seen in Figure 4, and they are Cheap_Dine_In,
Expensive_Dine_In, DineIn, Delivery, and NoPreference. Using these 6 global variables, each of
these variables was set to have a list. Each list in each variable was a list of restaurants.

Figure 4: ​The variables and lists

The “My Location On The Map” button is used to help a user determine their current location.
As seen in Figure 5, when the app initializes the location sensor is enabled. After the “My
Location On The Map” button is clicked the activity starter is called to use the location data
provided by the location sensor to pull up a map with the user's current location, allowing them
to input the restaurant and receive directions to it. Initially, the app was going to utilize the MIT
App Inventor Map feature to pull up a map with the location of the restaurant marked but not
provide directions.

Figure 5: ​The code to run the view my location function

The "Go to Website" button takes the user to the restaurant's Zomato page. As seen in Figure 6,
When the "Go to Website" button is clicked it calls the web viewer to open the restaurant's
Zomato page by using the join feature to join the first part of the link with the restaurant name
from the selected place label and the last part of the link. The replace function is used to change
the spaces that are present in the restaurant's name with hyphens so the link works.

4
 Figure 6: ​The code to run the view website button

The restaurant selection spinner is controlled through the use of several if-then statements, as
seen in Figure 7. These control which list the restaurant is selected from based on the selected
option from the spinner.

Figure 7:​ The code to run the spinner

Initially, the idea was to use a list picker and store the address of the restaurant along with its
name and service type.

5
RESULTS

Table 1 shows that users would use the application on average 3.3 times a week with a standard
deviation of 1.55 times a week. Users used the application on average 3.1 minutes per usage with
a standard deviation of 1.45 minutes. Based on a scale from 1-5 for Future Use, the average
rating was 3.4 with a standard deviation of 0.8. An average of 2.5 means that users may or may
not return to use the app. Anything less than 2.5 means that users are very unlikely to use the app
again. Ratings higher than 2.5 means there is a greater likelihood that the user will use the app in
the future. The initial application’s average rating is greater than 2.5 so that means users are very
likely to revisit the application again in the future.

Table 1​: Responses from users for the initial application

From the survey results of the initial application, more filter options were added to attempt to
decrease the amount of time needed to find a suitable location. Additionally, to increase the
number of times a user would use the application in a week a website loader was added for the
user to check reviews and pricing of the location.

The results from our final application survey showed that the application was successful since
the average showed better results. Table 2 shows that users would use the application on average
4.3 times a week with a standard deviation of 1.35, which is 1 time more than the initial
application. The increase in usage per week means the new features were able to convince users
to revisit the application more often. Table 2 also showed that users used the application on
average 2.8 minutes per usage with a deviation of 0.87, which is 0.3 minutes less than the initial
application. This indicates that the final application successfully helps the user find a location
faster than the initial application. Finally, the users were once again surveyed on how likely they
would use the application in the future based upon a 1-5 rating. The final application had an
average rating of 4.1 with a standard deviation of 0.83. A 4.1 means that users are even more
likely to use the application in the future since there was an increase of 0.7 from the initial
application. This shows that users liked the final application more than the initial application.

6
 Table 2​: Responses from users for the final application

In conclusion, the final application was more successful than the initial application. The number
of times used per week increased meaning users would use the application more. The usage time
decreased meaning users were able to find a location faster. Finally, the rating increased meaning
users preferred the final application over the initial application.

The survey data was collected by sending the .aia file and a link to the survey in the EGR100
Slack channel. After waiting several days the data were processed using Microsoft Excel and
creating data tables and scatter plots.

Figure 8​: Usage time per visit for respondents for the initial application

7
 Figure 9​: Usage time per visit for respondents for the final application

CALCULATIONS

Microsoft Excel was used to calculate the mean and standard deviation of the usage time and
how many times used per week for the final and initial application.
However, below shows how they would be computed manually without Excel.

Equations used:

Mean = Sum of terms/Numbers of terms

Standard deviation equation below:

𝛔 ​= sample standard deviation

N​ = the size of the sample
xi​ ​ =
​ each value from the sample
µ​ = the sample mean

Mean and standard deviation calculation of how many times per week the sample population
used in the first version of the app.

Mean = (3+6 2+1+4+3+2+5+5+2) /10 = 3.3 times per week

Standard deviation = ​𝛔 ​= 1.55 times per week

8
N​ = 10
xi​ ​ ​= 3,6,2,1,4,3,2,5,5,2
µ​ = 3.3

Mean and standard deviation calculation of usage time the sample population used in the first
version of the app.
Mean = (2+2+2+1+3+5+6+3+3+4)/10 = 3.1 minutes
Standard deviation = ​𝛔 ​= 1.45 minutes
N​ = 10
xi​ ​ =
​ 2,2,2,1,3,5,6,3,3,4
µ​ = 3.1

Mean and standard deviation calculation of how many times per week the sample population
used in the final version of the app.

Mean = (3+6+4+2+4 +3+5+6+6+4) /10 = 4.3 times per week

Standard deviation = ​𝛔 ​= 1.35 times per week
N​ = 10
xi​ ​ =
​ 3,6,4,2,4,3,5,6,6,4
µ​ = 4.3

Mean and standard deviation calculation of usage time the sample population used in the final
version of the app.
Mean = (3+2+2+3+2+5+3+2+3+3)/10 = 2.8 minutes
Standard deviation = ​𝛔 ​= 0.87 minutes
N​ = 10
xi​ ​ =
​ 3,2,2,3,2,5,3,2,3,3
µ​ = 2.8 minutes

CONCLUSION

The use of a cell phone application would help solve the problem of there being so many choices
of restaurants to eat out at and would help people who are struggling to find a restaurant to eat
out at. Through the use of the EatGreen app, students living on and near campus have a quick
and easy way to easily find restaurants to eat at around East Lansing based on the price and
service type of the restaurant. The time it took for users to find a location to eat was lower in the
final application, additionally with more choices the rating and usage of the application rose.

9
REFERENCES

1. B. Anderson, S. Lyon-Callo, C. Fussman, G. Imes, and A.P. Rafferty, “Fast-Food

Consumption and Obesity Among Michigan Adults,” Centers for Disease Control and
Prevention, Atlanta, Georgia, United States, Rep. A71, vol. 8, Date. Accessed: Oct. 11,
2020. [Online]. Available: ​https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3136980/

2. B. Okumus, F. Ali, A. Bilgihan, and A.B. Ozturk, “Psychological factors influencing

customers’ acceptance of smartphone diet apps when ordering food at restaurants,” Int.
Journ. of Hosp. Mngmt., United Kingdom, Rep. 7, vol. 72, Date Accessed: Oct. 8, 2020.
[Online]. Available:
https://www.sciencedirect.com/science/article/pii/S0278431916303103?via%3Dihub

10