Airlines' sustainability is improving:
Digital twin of in-flight catering operations
DESIGN FOR UNITED AIRLINES ( 2022.9 - 2022.11 )
INFORMATION DESIGN · DASHBOARD VISUALIZATION · PREDICTIVE ANALYSIS
Phygital and adaptable tracking.
Actionable insights and prediction.
INTRODUCTION
This digital twin is a multi-touchpoint data collection and analysis platform to track in-flight food consumption, and offer catering teams insights and plans based on cost, waste rate, and passenger satisfaction.
SPONSOR CORPORATION
United Airlines
BaoTran Le, UA Innovation lab
Anthony Haloulos, UA innovation lab
Jorie Sax, Catering manager
THE TEAM
PROBLEM
Severe food waste is due to UA's lack of measuring customer food preferences and consumption. They rely more on assumptions instead of data to make decisions on the food offered.
SOLUTION
Dashboard (core of the digital twin) :
(1) Tracking food consumption at a seat number granular level; (2) Comparing trends across different flights on the same route; (3) Actionable insight cards for predictive analysis.
IMPACT
-
Reported and presented to United's innovation lab, flight catering mngt team, and IIT smart tech lab.
-
It will dramatically contribute to UA's sustainable goal in 2050 along with the digital twin implementation.
-
The dashboard, as a core part of digital twin presence, improved its SUS score from 47 to 82.
C O N T E X T B U I L D I N G
UNDERSTAND AIRLINE OPERATIONS
3 Operators/Staff Interviews
Scope of airlines' ops
General processes, pain points, partners, and touchpoints in ops
5 Passenger Interviews
Experience in journey
Customer archetypes & behaviors, utterances, and satisfaction
Organization Study
UA ops' performance
Mission, current technology-in-use, challenges, and constraints.
We scoped into "in-flight catering management" by defining:
-
Target audience: UA in-flight catering management team
-
Stakeholders involved: Third-party vendors (ex. LSG), Crew staff, Passengers, and UA technology team.
-
Design goals: to tailor catering plans to achieve sustainability with more intelligence and less effort.
P R O B L E M S T A T E M E N T
UA in-flight catering management teams lack ways to track, measure, and quantify customer food consumption, wastage and preferences.
-
HMW phygitally enable United Airlines to track passengers' food consumption and sustainably manage wastage in a data-accumulative and predictive way?
UNDERSTAND DIGITAL TWIN TECHNOLOGY
20+ Literature Review
Digital Twin technology system mechanism
General processes to build and implement, required enablers
4 SMEs (Subject matter experts) Interviews
Constraints / challenges in IoT use cases
Smart connected product system related to digital twin
12 Case Studies
Similar applications in adjacent industries
DT in Logistics/inventory, Airport ops, Retail / Supply chain.
WHY DIGITAL TWIN - TECH ADVANTAGES
Data tracking, storage, and mapping of real-world touchpoints.
More mature implementation infrastructure than AR/VR.
Predictively analyzing and optimizing in virtual environment.
DESIGN PRINCIPLES
#1 Balancing automation techs and additional labor efforts or learning costs
#2 Providing easy access / benefits / incentives when requiring behavior changes
#3 Building scalable concepts with a data-developmental and adaptable perspective
#4 Always using United's terminologies and languages to smooth employee's onboarding
More research insights: please see this <Insight Book>.
A S S U M P T I O N S & H Y P O T H E S E S
HYPOTHESE FRAMEWORK
Proposed solution
We believe virtual kitchen to test passenger-profiles-tailored meals,
Success metrics
Will validate the gap between preferences and actual consumption,
Evidence & Data
Because ~47% of meal wastage is due to "unwanted meals".
STRATEGY AFFINITY MAP & CONCEPT COMBINATION
Concept prioritization details: please see this <Concept Book>.
SCENARIO + SERVICE BLUEPRINT
ACCEPTANCE CRITERIA
-
Digital twin touchpoints allow continuous input of sufficient multidimensional data to analyze metrics such as Meal completion rate, Food waste rate, Passenger satisfaction, etc.
-
It can track and compare data on multi-variables such as time, route and passenger load, and flight status.
-
Digital twin can provide managers with quantified and actionable insights and future plans.
T E S T & I T E R A T I O N
ITERATION THROUGH LEARNING
Sketch model
Build DT model
Design information hierarchy among the data ecosystem.
Living lab / Workshop
Measure the success
Prototype our concept and test in a participatory session with UA
Field visit and observation: LSG
Learn adoption barriers
How Digital Twin technology can be adopted in current scenarios.
DASHBOARD DESIGN | INFORMATION ARCHITACTURE TO HIGH-FI
Hover to see Information architecture iterations:
(1) Add more hierarchy to the data; (2) Highlight Insight section, make it more actionable
Feedback from workshop testing:
Digestibility - "Long lists and overwhelming data!"
➡️ Separate information in different sections and consider scroll depth.
Actionability - "What does this data mean to UA? What should I do next?"
➡️ Implements data prioritization with context, to ensure key insights' visibility.
F I N A L D E L I V E R A B L E S
DASHBOARD FEATURES
The trend chart shows the most important indicator of change over time: Numbers and rates of untouched meals. Managers can compare to the same time in the past to see why the data fluctuates.
Insights cards are ranked by Waste Rate, Cost, and Passenger Satisfaction, showing suggested changes in meal choices and their estimated improvements.
Showing data collected throughout the in-flight catering journey with structures, easy to filter, easy to track.
DESIGN SYSTEM
FUTURE DEVELOPMENT
We can add more intelligence using robots and sensors to empower future data generation in digital twin.
See more stories in our <Final presentation slide>
I M P A C T
Presented to UA innovation lab, catering mngt team, and IIT smart tech lab. United Airlines teams show expressed a strong interest in our project and are currently in the process of further handoff and implementation.
-
"We loved seeing the final products you come up with, both incorporating our interest in exploring newer tech while creating a human-centric design to help United's operations. " - BaoTran Le (UA Innovation lab)
-
In terms of Dashboard usability, after iteration and optimization, its SUS score was improved from 47 to 82.
DESIGN PROCESS
( Refer to Dan Nessler, Rethink design process, uxdesign.cc )