TrolleyFlow

It provides pick-and-pack operators with instant decisions on alcohol bottles, reducing effort and waste. It transforms the manual process into a smart and efficient operation.

Comment

Inspiration

Our inspiration was deeply rooted in the human aspect of airline operations.
We believe that the Pick-and-packer plays an essential role in the catering process, and we wanted to reduce their manual workload while improving efficiency for both the airline and Gate Group.
We also aimed to create a positive environmental impact by reducing waste and minimizing fuel consumption per flight.

What it does

TrolleyFlow is an intelligent operations system designed to optimize the management of bottles and liquids on flights.
The user (Pick-and-packer) selects the assigned airline and the number of bottles that must be prepared for a given flight.
Through the scanning of each bottle, the system automatically makes the final decision on whether to remove, discard, reuse, or refill it, eliminating the need for independent or subjective judgment by the operator.

How we built it

We built TrolleyFlow using a combination of tools and platforms:

  • A Figma interface as the prototype for our desired user experience.
  • Visual Studio Code for backend logic and system implementation.
  • Google AI Studio to integrate predictive models for forecasting bottle consumption and replenishment.

The architecture connects decision automation, scanning, and prediction modules into a cohesive, scalable framework for catering operations.

Challenges we ran into

The main challenge we faced was data collection and structure, as much of the process relied on assumptions and small internal dailies with the Gate Group team.
We also encountered difficulties in defining decision-making criteria for items that lacked a standardized database, and in integrating predictive analytics into a realistic operational environment.

Accomplishments that we're proud of

We are proud to have developed a functional and scalable solution that demonstrates real operational impact:

  • Improved efficiency for Gate Group operations.
  • Reduced waste in airline catering processes.
  • Automated decision-making that supports the Pick-and-packer role.

Additionally, we are proud that our system is extendable beyond alcoholic beverages to include other materials managed by Gate Group, such as napkins, plates, and even perishable food items.

What we learned

We learned how to use new tools, such as Google AI Studio, and developed a more critical and structured approach to assessing operational items.
We also understood the importance of building models with incomplete or limited data and how to generate reliable, data-driven decision processes in real-world scenarios.

What's next for TrolleyFlow

The next step for TrolleyFlow is scalability.
Our integration with the forecasting API allows the system to generate data-driven predictions on bottle usage and inventory rotation, providing high value for both the airline and Gate Group.
This will optimize:

  • Human capital costs for Gate Group.
  • Operational and fuel costs for airlines.

We envision TrolleyFlow as a realistic, sustainable, and scalable solution that can transform the future of intelligent airline catering operations

Built With

  • https://github.com/ate013lo/trolleyflow2025
Share this project:

Updates