Food Alarm

An attempt to reduce food waste amongst university students.

Type: User Research & Evaluation | Academic

Roles: Researcher | Designer | Developer

Duration: ~8 months (Oct 2017 - May 2018)

Tools: Java | XML | Google Opinions | Android Studio

Deliverables: Undergraduate Dissertation

How It Started

As part of my undergraduate degree in Computer Information Systems, I worked on a final year project concerning persuasive design and HCI. In essence, it documents the design process of an Android application that attempts to address food waste amongst fellow undergraduates. While the actual project involves theoretical research and coding, most of it has been omitted and adapted to focus on user research.

User Research

The project started with a vague notion to save the environment by encouraging fellow undergraduates to lead a sustainable lifestyle. However, a sustainable lifestyle encompasses many activities such as recycling, green transportation, efficient energy consumption etc. Therefore, a questionnaire was circulated to identify a specific domain to focus on.

The results suggested food waste to be the most suitable area since it was a less-known topic amongst the audience and potential for greater contribution.

Further research revealed an individual's shopping, cooking, and storing habits were the main sources of food waste (at a personal level). This informed the design of semi-structured interviews with the following key takeaways.

  • shopping behaviour varied but majority bought the same items every week

  • the main reason food went past expiry date was because they were unaware of what they owned

  • underlying motivations for avoiding food waste included loss of money & guilt

Defining the problem

"To help students consume all their purchased food by making them aware of what they own"

Personas

A couple of personas were created using the research data to capture the essence of the target audience. Although not comprehensive, these provide a quick overview of the target audience.

A persona named James who throws away food because he does not remember what he owns
A persona named Erin who purchases groceries in excess

Ideating solutions with low-fidelity prototype

As mentioned earlier, money and guilt were the dominant reasons why people felt bothered about food waste. Therefore, the positive spectrum of these aspects were considered as motivation while ideating solutions along with the problem statement.

hope of saving money instead of fear of losing money
hope of saving the environment instead of fear of harming the environment
pleasure from performing good actions for society

The project requirements along with my experience and research led to the solution in the form of an Android application. Low fidelity paper sketches were shown to prompt the audience to get their initial feedback and opinions.

Low fidelity early concept sketches in mobile sized papers with various interfaces

Early user feedback

  • Disliked the idea of a tree growing as a result of more sustainable actions

  • Favoured the concept of separating wasted food and consumed food

  • Wanted more practical features e.g. history, performance

  • Wanted it to be more engaging and fun

  • Wanted to simplify mechanism to save a food item

Ideally, I wanted to spend more time exploring different possibilities but had to move on to prototyping due to time constraints.

High-Fidelity Prototype

The feedback from the sketches was incorporated into the following high-fidelity prototype. The aim was to develop a functioning prototype with key features for evaluation purposes.

Navigation structure


                Navigational structure with the home screen at the top which leads to add item, delete item, achievements, learn, help and 
                settings screen. Efficiency, badges and history screen can be accessed from achievements screen.

How it works

1. Add

The user can add food items they purchase to the list including its name, type and expiry date.

2. Throw, Consume

The user can remove the food item from the list by selecting whether they consumed it or threw it away.

3. Monitor

The user can monitor their performance in the achievements screen, which shows the proportion of food they consumed.

Individual screens

Includes a card with user details
                that have game-like attributes at the top. Two adjacent cards indicating consumption 
                and money saved. The list of saved food items are placed at the bottom.
Add food features three input fields - the
                food's name, type and its expiry date.
Features a 
                dialog to ensure user confirmation.
Displays the percentage of the food the user has consumed, and the 
                amount of money saved compared against the national average.
Displays the badges the user has unlocked according to their performance.
Displays 
                the amount of food the user has registered, thrown, consumed and the type of food the user has thrown 
                out the most.
Displays information regarding
                how to use the app.
Displays brief information on
                harmful effects of food waste on the environment.
Enables the user to 
                change the display name in the home screen and set their preferred notification time.

User Evaluation

To measure the effectiveness of the persuasive features in the prototype, it was sensible to compare it with another similar product without such features. Therefore, Expiry Wiz was chosen as the control system.

Experiment Design

A between-subject approach was implemented where 12 participants were pre-screened and split into two groups of 6. There were three phases in the experiment:

Phase 1

Pre-study questionnaire to establish a baseline & task-based usability test to assess efficiency and effectiveness.

Phase 2

Daily use of their respective applications for four days to expose them to the concept and gather opinions.

Phase 3

A post-study questionnaire with System Usability Scale (SUS) to measure usability & interviews for insights.

Usability Results

Food Alarm ExpiryWiz
Average time taken for task 1 * 167.17s (faster) 201.67s
Average time taken for task 2 ** 77s 60.83s (faster)
Completion 100% 100%
SUS score 89.2 (better) 70.4

* Task 1 involved adding a list of food items in the app

* ** Task 2 involved removing specific items off the list

 
In summary, usability was measured through the SUS scale and the task-based usability test. Although there was no difference in task completion rate and minor differences in time taken to complete the tasks, there was a significant difference in their SUS scores.

Insights from post-study interview

  • 5 of 6 participants claimed that they would not use Expiry Wiz again while everyone in the Food Alarm said they would use it again in the future with minor improvements.

  • Participants using Food Alarm were more motivated to perform the tasks because they could receive extra information and satisfaction from the process and effort.

  • The inclusion of gamification features showed mixed reactions. A few did not realize that the feature was available.

  • A few were confused in regards to the figures in the home screen, suggesting labels wherever needed.

  • The user details which were provided for self-monitoring purposes in the home screen did not provide any value to the users.

  • Both groups expressed the need for a better logging mechanism and provided suggestions such as a barcode scanner or photo recognition system, which would automatically register the food item into the list.

 
In general, participants from the Food Alarm group showed more enthusiasm towards the system suggesting the prototype to be more fun and useful.

Reflection

There were many aspects the project could have improved on - time management, project management, resource allocation, etc. As a result, I was unable to do everything that I wanted to do and compromises had to be made.

In particular, I learned that evaluation requires a lot of time and thorough planning. Recruiting real target users can be time-consuming as individual schedules need to be considered on top of suitability. Plus, participants may drop out mid-way which need to be addressed beforehand.

Nevertheless, this project was worth all the sleepless nights. Not only did I learn the theoretical aspects of persuasive design and HCI, but I also really enjoyed the practical aspects of the process which turned my curiosity of being a UX designer into an aspiration.

Other selected work