SIWAN KIM

Pelli: Drive-Thru UX with Autonomous Delivery Robots

Mar 2020 - Jun 2020

Summary

I led the design of Pelli, an autonomous robot-based drive-through service aimed at enhancing user convenience and safety. To refine the service, we conducted in-depth interviews to gather insights, which guided the creation of personas, user journey maps, and service blueprints. Key features were prototyped based on these findings. The project received the highest evaluation in class, was further developed through a startup camp and idea competition at the KAIST Eco-friendly Smart Vehicle Research Center, and won an award. Additionally, it was later presented as a poster at HCI Korea 2021.

Proposal of Autonomous Robot-Based Drive-Through Service, Proceedings of HCI Korea 2021

My Role

User Research
UX Concept
Prototype Design
Team Leader

Course

UX Design Portfolio I
Graduate School of Information, Yonsei University

Team

Master students : 3

Problem Definition

How might we reduce the complexity of drive-thrus
while enabling users to pick up a wider variety of food options?

We conducted in-depth interviews with adults in their 20s and 30s who had experience using drive-thru stores (DT).
Through two rounds of analysis on 286 raw data points, we identified 3 key insights.

Reducing Time Spent at Drive-Thru Stores
Users want faster and more efficient service, particularly in reducing waiting times during ordering, payment, and product pickup.
Simplifying In-Car Ordering and Payment
Users need a smoother process for browsing menus, placing orders, and making payments within the car, minimizing interruptions or confusion.
Improving Product Delivery Efficiency
Users value seamless handoff processes that ensure accuracy and reduce delays when receiving their orders at the pickup window.

The Approach

Over four weeks, we utilized various UX methodologies, including stakeholder mapping, value proposition mapping, personas and user scenarios, user journey mapping, and service blueprints, to explore solutions to the identified problem. Through this process, we clarified how the service we provide delivers tangible value to our users.

UX concept

As-is
To-be
Impact

As-is
Users wait in line or at a parking spot for their food.
To-be
Users are greeted by a robot that ensures precision and speed. This redefines convenience with an innovative and seamless handoff experience.
Impact
You don’t have to wait — Robot Pelli is there for you.

As-is
Users can only order from one store at a time
To-be
Users can Bundle multiple orders from different stores in one trip, expanding their options for a personalized and diverse meal.
Impact
You don’t have to compromise — Order from multiple stores at once.

As-is
Users must set a specific drive-thru location as their destination.
To-be
Drive-thru stops seamlessly fit into the user's journey, making it part of their routine rather than a detour.
Impact
You don’t have to guess — Plan your journey with precise time predictions.

Prototype

Set Destination & Departure

Users enter their route and departure time, and the system recommends the best drive-thru locations using real-time traffic, vehicle, and store data.

Explore Available Drive-Thru

Discover nearby drive-thru options with a curated list of restaurants and retail stores. Easily check the additional travel time required for each location and make an informed choice.

Order & Pay

Add items to your cart and create your own combination from multiple stores within the same marketplace. Save time by completing your payment before you start driving.

Review your order while Boarding

Check your order details and make last-minute changes before finalizing it.

Real-Time Updates and Pickup Zone Navigation

Receive real-time voice updates for cooking progress, completion, and the movement of Pelli. As you approach the pickup zone, easily locate its exact position.