SEO UX DESIGN RESEARCH LAB
Graphic Design and Media, B.S., University of Nevada, Las Vegas (UNLV)
Research group: Kaylin Joung, Yuki Hayashi and Dailuaine Esguerra
Research accomplishment: 
- Spring 2025 Undergraduate Research Symposium (Research Presentation)
​​​​​​Title: Designing AI-Driven Dining: A UX Approach to Enhancing the Self-Service Experience
Research Keywords: User Experience (UX), Artificial Intelligence (AI) in Dining UX design, Food Service Robot Design, Smart Restaurant Experiential Technology
Abstract: As technology has rapidly advanced in recent years, the integration of artificial intelligence across various sectors has become increasingly prevalent. Technological progress has led to various AI-driven mechanisms aimed at enhancing self-dining services through faster service, greater convenience, and improved accuracy. By leveraging AI, this study examines how dining technologies can significantly elevate the overall dining experience in All-You-Can-Eat restaurants in Las Vegas. This research focuses on developing UX design concepts for a hypothetical self-dining scenario and explores how the integration of AI language translation, food server robots, a self-service kiosk system, and a voice interaction system can enhance the dining experience in the specified scenarios. The research methodology began with an initial survey to analyze users' dining experiences, followed by the development and visualization of design concepts — including user flows, storyboards, branding, high-fidelity wireframes, and design prototypes — to support the successful adoption of AI robots. We crafted user-centered narratives to envision an immersive dining experience with a prototype design, 'Dishie,' integrating AI-powered language detection for smooth, natural communication. Our findings emphasize the importance of balancing automation with human interaction while prioritizing user privacy. By offering a customizable ordering system, Dishie enhances convenience without compromising personal choice. These findings suggest that thoughtful AI integration, when paired with human-centered design, can enhance operational efficiency while preserving meaningful customer interaction and personal choice.
Research Objective: This research explores how the combination of existing technologies such as AI language translation, food server robots, self-service kiosk systems, and voice interaction systems can be integrated into the dining experience to enhance the general dining experience, while ensuring more inclusive and seamless experiences for all customers, regardless of their status as a first-time, tourist, or a returning customer. 
Research Question and Hypothesis: We hypothesize that our kiosk design can achieve this by providing intuitive visual/audio guidance, automatic language adaptation, and incentivizing return visits through loyalty rewards and personalized recommendations based on past dining data. To test this hypothesis, our team will design and develop AI-integrated at-table kiosk and robot prototypes that prioritize usability, language accessibility, and personalization. By analyzing industry trends and user needs, we aim to create innovative solutions that transform the self-service dining experience for diverse customer groups.
Methodology: We conducted an autonomous online survey to gather insights into user preferences and expectations for self-dining services powered by AI. The survey, drafted in Google Forms, included 20 questions and received 32 responses from participants aged 18-55. Notably, 53.1% of the respondents were aged 18-24, a key demographic for innovative dining experiences. Most participants reported dining out either a few times a month or once a week, making them well-acquainted with restaurant environments. Key findings revealed that popular AI tools such as ChatGPT and Google Gemini are among the products they enjoy most. Additionally, "fast, convenient, and easy to use" emerged as the top features users appreciate in translation tools or similar methods they currently use. When it comes to AI-powered self-dining experiences, respondents emphasized two critical features they would like to see: ease of use and the option to seamlessly switch to a human assistant if needed. These insights highlight the importance of combining efficiency with flexibility to meet diverse user needs in self-dining scenarios.
User Flow Chart: This user flow chart was created to illustrate the step-by-step process the user takes during their dining experience powered by Dishie, ensuring all the actions that we intend the customer to take are covered– from first encountering the self-ordering kiosk to potentially signing up for a loyalty membership before their departure from the restaurant. It then served as a guide for translating these flows into a storyboard and functional prototypes on Figma.
Storyboard: The All-You-Can-Eat restaurant integrates AI-powered self-service technology to enhance the dining experience through efficiency and personalization. Customers order in their own language via voice-activated kiosks or smart speakers, depending on their seating. Orders are delivered by a robot named Dishie, and payment is completed through the kiosk, with an option for human assistance if needed. The experience combines automation with convenience, and a loyalty program encourages return visits.
Improving dining experience with AI-powered food server robot and interactive table interface: When the customer enters the restaurant, they will be greeted by one of the robot servers (1), who will then prompt the customer to enter information like whether they made reservations online or the number of guests (2). Based on the information, the server robot will guide the customer(s) to a table that best suits their needs (3). Once seated, the interactive table and an audible instruction will guide the customer to utilize the touch screen of the interactive table to start their order (4). The touch screen can display multiple digital menus to match the number of customers seated and allows each customer to make individual orders at any point (5). The digital menu also allows customers to select and open a detailed description page for each menu item, where a life-sized and detailed preview of the dish can be viewed before finalizing the order (6). After completing their order, the interactive table will have different entertainment applications or a customizable digital tablecloth option that can be changed while they wait for their order to arrive (7, 9). Once the order is ready, the server robot will make its way to the customer’s table with their orders. To clarify which food is being delivered to the customer, a green light indicator will light up on the corresponding tray on the server robot (8). When there are empty dishes, they can be placed in the designated area on the table, where sensors will detect the empty dishes (10). When empty dishes are detected, a retrieval robot will be summoned to the table, where the customer can simply place the empty plates. Once the customer is ready for payment, they can choose to call a human server to handle the payment process, or they can make contactless payment directly on the table with the integrated Near Field Communication (NFC) pad (11).
Design Findings: In developing the AI-assisted dining experience, the design process focused on creating flexible solutions tailored to diverse restaurant needs. Two primary models of the Dishie system were proposed: one with a movable tablet equipped with microphones and robotic food delivery, and another featuring a fixed interactive screen. These options provide adaptability based on restaurant scale and budget. The design emphasized a thoughtful integration of automation and human interaction, ensuring that staff could intervene when necessary to maintain a sense of personalized service. Additionally, Dishie was designed to support optional user accounts that store order history and dietary preferences, enhancing convenience while prioritizing user privacy and data control. This dual approach offers both operational efficiency and a customizable, human-centered dining experience.
Prototype Concept: The design of the robot functions as a carrier for food, with multiple shelves holding small plates. The robot's face is a screen pad that displays the Dishie logo when idle and shows the table number it is serving when active. Lights on the side of the shelves glow green or red, green indicating that the food is ready for the customer to take, and red signaling that the item is designated for another table. The self-ordering touch-screen model presents the user interface design. It is not fixed to the table, allowing flexibility in its placement around the dining area. The touch-screen table enables customers to place orders directly through the surface, allowing them to browse the menu without needing a shared booklet or kiosk tablet. A circular dual microphone and speaker, positioned at the center of the dining table, captures voices and projects sound in all directions, providing 360-degree audio coverage to facilitate seamless voice interaction.
Conclusion: Our research highlights the significant impact of artificial intelligence on self-dining services, particularly in enhancing speed, efficiency, entertainment, and customer convenience. One key takeaway is that the integration of smart speakers and kiosks should consider environmental factors, such as the need for louder volume in open areas. Furthermore, continuous AI upgrades are needed in order to hinder the efficiency of self-dining automation. Consistent updates and refinements are essential to maintain up-to-date users’ needs and to address current limitations in AI-powered systems. Lastly, real-world pilot testing and usability testing are key to identify practical challenges, user feedback and behavior. There is still plenty of room for exploration in Dishie’s development. The next step is to conduct another round of surveys or in-person interviews with restaurant owners. This research will focus on their experiences with non-English-speaking tourists, the challenges they face from both the customer and business perspectives, and their overall impressions of Dishie, including whether they would consider implementing it in their restaurant. Working with an engineering team will also be important to identify technical constraints. Future research should look into the challenges of turning Dishie from a design concept into a functional prototype, especially as AI and language detection technology continue to advance. Finally, in-person testing with a working prototype should take place in a real restaurant setting. A popular all-you-can-eat restaurant that attracts international customers would be an ideal location to temporarily implement Dishie and gather valuable user feedback.