PROJECT PART 4 - PROTOTYPING & EVALUATION

PROJECT PART 4: PROTOTYPING & EVALUATION

Subject: Human-Computer Interaction (SECV2113)

Session: 2024/2025 Semester 2

Lecturer: Ts. Dr. Sarina binti Sulaiman

From: Group 7 Powerpuff

Group leader: Shannon Toh Jia Ee

Member:

1.     Ng Esther

2.     Kam Kai Xin

3.     Chong Wen Hui

 

A.    Introduction

GreenSort is a smart waste sorting system developed to support responsible waste management in urban and commercial environments. Designed with an AI-powered avatar, GreenSort helps users identify and dispose of waste items accurately through an interactive smart bin interface and a connected mobile application. The system also supports municipal operations by offering real-time bin monitoring and waste reporting features, aligning with key principles of user-centered design.

The main goal of this project phase was to evaluate the usability and functionality of the high-fidelity prototype built using Figma. This prototype focuses on three core user tasks:

1.     Sorting waste using the smart bin interface

2.     Monitoring bin fill levels via the mobile app

3.     Generating and viewing a waste report

Usability testing was conducted to gather feedback from real users representing our three main target groups. The feedback helps us identify usability issues, improve the interface, and validate the interaction flow of the system.

Testing responsibilities:

Testing for User 1 (Miss Celeste – Municipal Waste Worker) was conducted by Kam Kai Xin

Testing for User 2 (Miss Rachel – Urban Resident) was conducted by Shannon Toh Jia Ee

Testing for User 3 (Mr. Ng – Facility Manager) was conducted by Chong Wen Hui

 

B.    Screenshots of the prototype

1)     HOME PAGE

 

2)     SMART BIN INTERFACE (REFER TASK 1)

 

3)     BIN FILL LEVELS (REFER TASK 2)

 

4)     WASTE REPORT (REFER TASK 3)

 

C.    Briefing notes – prepared by {Ng Esther}

{Hi, thank you for agreeing to test our system. GreenSort is a smart waste sorting bin system that uses AI to identify waste types and guide users in disposing of them correctly. The system includes a mobile app for checking bin status and viewing waste reports. You will be asked to complete 3 tasks. Please think aloud while doing each one, so we can understand your experience. If you feel stuck or don’t know what to do, you may say “terminate” and we’ll move to the next task. We’re testing the system, not you—there are no right or wrong answers.}

 

D.    Testing with users

 

[1]   Task 1: Sort Waste Using Smart Bin {by Kam Kai Xin}

User 1 video link: https://youtube.com/shorts/-qZ5rJRZ-WQ?feature=share

User 2 video link: https://youtube.com/shorts/pZANMsKIPOA?feature=share

User 3 video link: https://youtube.com/shorts/Zer9ozqBs_Y?feature=share

 

[2]   Task 2: Monitor Bin Fill Level {by Shannon Toh Jia Ee}

User 1 video link: https://youtube.com/shorts/V4ViuW07Ct8?feature=share

User 2 video link: https://youtube.com/shorts/FmggSEHU-ic?feature=share

User 3 video link: https://youtube.com/shorts/aIispfD-AUU?feature=share

 

[3]   Task 3: Generate/View Waste Report {by Chong Wen Hui}

User 1 video link: https://youtube.com/shorts/O39FUPmFQy8?feature=share

User 2 video link: https://youtube.com/shorts/7wU_YTFLLA8?feature=share

User 3 video link: https://youtube.com/shorts/p_LkVNM-ACU?feature=share

 

 

E.    Observations – prepared by {Ng Esther}

Throughout the usability testing sessions with Miss Celeste, Rachel, and Mr. Ng, we observed how each user interacted with the GreenSort system while completing the three tasks: sorting waste, monitoring bin status, and generating/viewing reports. The following summarizes our observations and interview feedback:

Miss Celeste initially approached the system cautiously but responded well to clear visual and voice guidance. She successfully sorted the item using the scanner and understood the color-coded bin indicator. However, she showed signs of hesitation during the report download task due to small font sizes and the lack of confirmation messages after pressing buttons like “Schedule Pickup.” She suggested using larger text and confirmation prompts to improve her experience.

Rachel moved through the tasks smoothly and was particularly engaged with the Eco Stats feature. She was comfortable with the interface layout, especially the use of icons and graphical charts. She appreciated the ability to scan or use voice input to sort waste but noted that first-time users may not immediately know which scan option to choose. During the interview, she recommended a guided walkthrough or tutorial mode for onboarding. Rachel also liked the reward badges, stating they made the experience more enjoyable.

Mr. Ng focused more on the functionality and practicality of the system for facility management. He successfully navigated the bin map and was quick to locate which bins were full. However, he expressed confusion regarding the meaning of “Offline” status and wanted more details on whether this indicated a serious issue. In the reporting section, he managed to download the report but found the file name generic and the data layout a bit unclear. He recommended clearer bin labelling and a summary view for quick reference.

In conclusion, all users completed the tasks but expressed areas where usability could be improved. Visual clarity, feedback mechanisms, and additional guidance were the key themes across all observations. These insights helped inform the design refinements for the next phase.

 

 

F.     Findings – prepared by {Ng Esther}

 

From the usability testing of the GreenSort prototype, several interface-related usability issues were identified through observation and user feedback tied directly to the screens presented:

1. Text Readability in Bin Status Dashboard

In the first screen showing "Your Smart Bin Status," some users (particularly Miss Celeste) mentioned that the percentage values and bin labels were too small to read comfortably, especially when used under outdoor lighting or while wearing gloves.

 Proposed Solution:
Increase font size of key values (e.g., “65% full”) and use bolder contrast between background and text for easier visibility.

 

2. Color Dependence on Map View

In the “Bin Status Map” screen, color alone is used to indicate bin status (e.g., red = full, green = available). While Rachel and Mr. Ng could interpret these quickly, some concern was raised about accessibility for colorblind users or those unfamiliar with the color coding.

Proposed Solution:
Add supporting labels (e.g., “Full,” “Offline,” “Empty”) next to or under each colored dot. Also consider using patterns or icons to supplement color meaning.

 

3. Lack of Confirmation for ‘Schedule Pickup’ Button

The large green “Schedule Pickup” button is prominent and clear. However, during testing, users were unsure whether tapping it actually triggered a pickup request, since there was no confirmation message or animation.

 Proposed Solution:
Include a confirmation pop-up or toast notification (e.g., “Pickup request submitted!”) with success sound or checkmark animation.

 

4. Unclear Achievement System in Eco Stats

On the “My Eco Stats” screen, the achievements (e.g., Recycling Hero, Compost Master) caught Rachel’s attention and encouraged her to stay engaged. However, she commented that it wasn’t clear how to earn them or what the point system meant.

 Proposed Solution:
Add a “Learn More” link or tapable tooltip icon beside each achievement that explains how it's unlocked and what benefits it brings.

 

5. Offline Bin Status Icon Confusion

In the map view, one bin is labeled “Offline” with a grey dot. Mr. Ng noted that while this likely means the bin is disconnected, the system doesn’t explain what that status entails or how urgent the issue is.

 Proposed Solution:
Include a tooltip or tapable label like “Offline – check connection or power” and highlight any urgent actions needed for admins.

 

6. Limited Task Guidance for First-Time Users

While Rachel navigated the app easily, she noted that first-time users might not know how to use features like “Scan Barcode” or “Voice Search” under the Waste Sorting Assistant.

 Proposed Solution:
Add brief text hints below each function (e.g., “Tap to scan product barcode”) or a simple onboarding screen when the app is launched for the first time.

These findings help inform specific interface improvements based directly on the tested prototype visuals. Addressing them will enhance the system’s clarity, accessibility, and user confidence across different user groups.