AI-driven Warehouse Operations Platform
AI-driven Warehouse Operations Platform
.Client
Alibaba Group
.My Role
UX Design, Product Design.
.Our Users
Support Associate, Order Picker, Inventory Control Specialist.
.Timeline
06.2024-09.2024, Launched 11.2024
.Team
YueYang - PM
BaoWang - Product Leader
QianNing - Engineer
ShiLang- Engineer
Rengqi Han- Data Scientist
.Overview
Warehouse operations were inefficient, with field efficiency constrained by outdated workflows on the digital platform. To address this, Alibaba sought an AI-driven solution leveraging digital interactions to boost productivity. This smart warehouse platform empowers workers, enhances efficiency, and lowers warehouse costs.
01
Context
Reimagine warehouse operations with a system that’s intuitive, efficient, and seamlessly integrated.
02
Problem
To build domain knowledge and identify key challenges, opportunities, and relevant stakeholders, we carried out 29 on-site interviews.
Through interviews with workers in various warehouse roles, we prioritized key issues and identified that the inefficiencies faced by Inventory Control Specialists, Order Pickers, and Support Associates can be effectively addressed through a platform-based system.
Inventory Control Specialist
Order Picker
Time-Consuming Item Recovery
Support Associate
The problem is encountered by all three roles
PC Window Overload
03
Field Research
To understand the core reasons behind these problems, we conducted a deep dive into field operations.
I led in-depth interviews with 9 users across three key roles—Support Associate, Inventory Control Specialist, and Order Picker.
Due to the unique nature of on-site operations, it was difficult to uncover all key issues through communication alone.
I facilitated task-driven research immersive roleplays to observe real operational habits of three key stakeholders, formulate hypotheses about the core probelms, and develop initial design solutions. This approach provided actionable insights for further refinement.
Insights
01
We found slow item picking is impacted by Inefficient routes and fixed visual focus which caused Pickers to overlook items in less prominent areas.
02
We found time-consuming item recovery is caused by Improper Picking and lack of information.
03
Communication is challenging in large warehouses, but most issues can be resolved through standard guidelines. High employee turnover leaves many unfamiliar with these procedures.
04
The PC window is overloaded due to multiple pages being opened and frequently referenced.
How to win strategy
01
From habitual gaze to intentional redirection
02
03
From time-consuming clarifications to instant AI insights
04
04
User Test
To validate hypothesis and assess whether the features were suitable for warehouse operations, I conducted user tests prior to creating the wireframe.
Research Question and Metrics: How much distance can optimized routes cut? How much time can location prompts save? Is voice input suitable for warehouses?
Findings
01
Optimal and inefficient routes differ by over 500 meters in running distance.
02
Reducing visual scanning significantly accelerates item searching.
03
Voice input is a suitable modality for warehouse operations.
05
Wireframe
Based on the research insights above, I developed a solution strategy and outlined the core product MVP, which led to the creation of the wireframe.
Cross-platform synchronization for consistent user experience between mobile and PC interfaces.
06
Retrospective
I facilitated three retrospectives with experts from different fields — UX, Data, and PM.
I presented my research findings and design solution, aiming to design from the core problem. However, engineers cited limited resources, suggesting starting with simple features while anticipating AI product execution next quarter.
Takeaway
Reschedule Design Hierarchy
We prioritized addressing core issues while balancing resource constraints, opting for a phased approach: optimize with split-screen support and design improvements, then redesign with intelligent features, and finally explore conceptual innovations.
07
Iteration
Solution 1 - Optimized Interface
I redesigned Interfaces and optimized workflow for on-site item retrieval.
Eye-Level Guidance: Enhances navigation by quickly pinpointing specific shelf positions.
Optimal Path Planning: Reduces detours with efficient pick routes and order sequencing.
Solution 2 - AI filter
I designed a cloud voice library that enables pickers to upload anomaly reports via voice, accessible to associates on PC and mobile. AI search efficiently retrieves clues, transforming individual guesswork into collaborative problem-solving .
Fast and Accurate AI Filtering: AI quickly identifies relevant clues, improving efficiency.
Field Access: On-site voice library access enables real-time adjustments for better item retrieval.
Solution 3 - Gen AI
I designed a mobile version of ChatAI, extending its functionality from the PC.
Flexible Input: Supports voice, shelf QR codes, and text as prompts, saving input time.
08
Integration
To enhance ChatAI's capability as a versatile internal tool for various business units, I addressed integration challenges across teams.
I collaborated with technical teams to develop a universal design pattern but faced issues like data security concerns and siloed management between overseas and local teams.
To resolve this, I designed three card solutions: a comprehensive option requiring broader data permissions and collaboration, and a basic option for teams preferring minimal resource and data involvement.
The three bundles offer flexibility tailored to team preferences and resource commitments.
09
Takeaways
A framework for field operation AI products
focusing on addressing on-site challenges with scalable and efficient solutions.
