UX Research/UX Design
KitchenBot
%20(1).png)
.png)
.png)
This project focuses on helping students in Ithaca who regularly buy and store fresh produce, which often spoils quickly. The goal is to reduce food waste and save money by providing a system that helps users track, manage, and use their groceries efficiently. Our solution, KitchenBot, combines a kitchen-based smart speaker with a companion app to enable inventory tracking, smart data input, and proactive reminders. Users can log groceries through photos, receipts, or voice commands, receive recipe suggestions based on available ingredients, and get alerts for items nearing expiration. By preventing spoilage and over-purchasing, KitchenBot helps students save money while making more efficient use of their food.
For more details, feel free to check out my medium article or reach out to me at yp2859@nyu.edu
Upperclassmen Cornell students who often cook at home need a method to reduce excess food waste, because they want to save money and not let food spoil.
The goal of the interview study on grocery shopping and cooking habits is to gain a comprehensive understanding of individuals’ behaviors, preferences, and challenges related to cooking activities. More specifically, in relation to the challenges around food spoilage.
Adam is a 21-year-old senior at Cornell University who does not have a meal plan and prepares his own meals.
The interview was conducted in Adam’s living room and kitchen at 1:00 PM and lasted approximately thirty minutes. Although Adam lives with two roommates, neither was present during the interview.
Jane is a 21-year-old senior at Cornell University who does not have a meal plan and prepares her own meals.
The interview was conducted in Jane’s dining room and kitchen at 5:00 PM and lasted approximately thirty minutes. Jane lives with one roommate, who was not present during the interview.
Students have a hard time figuring out how much food to buy when grocery shopping.
Students do not usually use an app or other methods to track their grocery shopping or cooking.
Students tend to go grocery shopping at least once every two weeks.
Students tend to eat out for convenience and social reasons.
Below are some of the ideas from my individual brainstorming session that I shared with my team as we refined and combined concepts into our final design. We evaluated each idea’s strengths, weaknesses, and relevance to the problem of fresh produce spoiling quickly. By identifying overlapping themes and complementary approaches, we were able to synthesize a solution that was both innovative and distinct from existing market options.
To help users reduce food waste, our team decided on the solution of an app and speaker interface called KitchenBot, with the speaker element situated in the users’ kitchen or wherever they prepare meals. It synthesizes concepts around inventory tracking, smart data input, and proactive reminders by allowing users to log groceries through photos, receipts, and speech, then using that data to suggest recipes, prevent over-shopping, and flag items nearing expiration.
Using the ten usability heuristics, our team evaluated our low-fidelity prototype. We each individually assessed the design, as heuristics help highlight potential usability issues early in the process. To the left is an image of the spreadsheet we used to collect data, detailing the problems encountered, the heuristic violated, a description of the issue, the severity of the violation, the suggested solution, whether a change should be made, and the justification for the decision.
Heuristic Notes
Each team member conducted user testing with one participant, for a total of four participants. To the right is an image of the spreadsheet the team and I used to organize problems observed during usability testing. It includes columns for location, problem description, evidence, severity, suggested solution, solution cost, whether to make the change, and justification for the change.
Usability Test Notes
In the previous iteration, the app faced several usability issues. Interactivity was limited because users could not manage reminders, follow or filter recipes, or confirm scanned items, which violated the visibility of system status heuristic. To address this, we added real-time indicators: reminders decrease when checked off, the cart updates as items are added, and scanned items are confirmed in the pantry.
Users also struggled to access KitchenBot’s voice interface, which violated the user in control heuristic. Originally, the robot icon controlled device settings, causing confusion. In the updated design, settings were moved to the user profile, and the robot icon now activates the voice interface, allowing users to give instructions, get recipes, and manage their digital pantry.
Finally, navigation and consistency issues such as dead ends on the login page, unwanted transitions, and missing back buttons were resolved by making the sign-in button tappable, removing unnecessary functionality, and adding back buttons, creating a smoother and more predictable user experience.
We created a video demonstrating how KitchenBot could be used in a real-life scenario, highlighting both the app and the speaker interface.
What went well:
If I were to do this project again, I would have:
