AI Scheduling
An assistive AI-integrated scheduling tool for the Mobility Air Force.
Case Study
Role
Product Designer
(Product Design Lead for scrum team)
My Contributions
Design Research
User Interviews
Research Synthesis
Interface and Interaction Design
User Workflow
Prototyping and Testing
Product Strategy
Visual Design
Agile Methodologies and Roadmap Prioritization
Award + Publication
2023 R&D 100 Award for Puckboard
Applying Human-Centered Design to AI-Enabled Pilot Scheduling
22nd International Symposium on Aviation Psychology.
Alexander, A. L., Haque, A., Snyder, M., Kusiak, R., Okubo, B., Chung, K., & Robinson, E. (2023)
Overview
The RVCM team (now acquired by Intellibridge in 2023), collaborated with MIT Lincoln Laboratory to optimize Puckboard’s scheduling tool through recommendations to calculate optimal aircrew for events while considering aircrew currency, availability, and burden distribution across the unit. This is an ongoing product and initiative.
Note: Some details have been omitted due to the nature of the work.
Challenge + Primary Audience
Assist Schedulers and Aircrew members within the Air Mobility Command’s C-17 Airframe by providing recommendations for optimal schedule and aircrew.
How can we build trust with the USAF C-17 community to utilize this recommendation tool as an assistive tool rather than a replacement of the scheduler role as well as to alleviate the many pain points for Schedulers and Aircrew?
Team
This initiative includes a large collaborative team between three entities made up of roles such as:
MIT Lincoln Laboratory
- Principal Investigator
- Human Factors Researcher
- Experience Design and Researcher
- Algorithm Engineers
United States Air Force
- Mobility Officer
- AMC Chief, Systems Branch, Requirements Manager
- AMC GTIMS Requirements Manager, Systems Branch Chief
- Various Aircrew Members
Puckboard/RVCM (now Intellibridge)
- Program + Project Managers
- Product Designers
- Tech Lead
- Software Engineers
- Subject Matter Experts
Research:
Workflow + Personas
Understanding Scheduler’s current workflow of scheduling without AI assistive tools within the Puckboard application through various user interviews.
Key Problem Statements:
As a Scheduler, I need to assign optimal aircrew members to events in Puckboard so I can save fuel and distribute burden effectively across my unit.
As a Scheduler, I need to view recommendations so I can assign optimal aircrew members to events in Puckboard.
As an Aircrew member, I need to view upcoming events and currency status so I can request events that will fulfill requirements to stay current.
Constraints & Opportunities
Building Trust: adjust to “Recommendation” vs AI language. Communicate that this is a tool to enhance scheduling rather than a replacement of the scheduler role.
Wait time: some participaints were resistant to the solver tool because it took a few minutes due to the limitations with ARMS and Platform 1 (USAF databases).
Key Takeaways + Insight
The following features would be helpful for participants:
Individual aircrew dashboard
Unit level (scheduler role) dashboard
Upcoming events
Aircrew recommendations
Event detailed currencies
Flight hour views
Legality constraint data
DNIF status
Impact statements
Design Process + User Feedback
Identifying key features, I reached out to users via 1:1 interviews through zoom and Mattermost to get feedback to build out the smart recommendation tool to iteratively improve designs and user experience.
Next Steps:
This product and Initiative is still ongoing with a new design system as well as new features to continue to improve and assist with the complexity of scheduling for the USAF C-17 community and we are in the process of expanding to other airframes.