Rerouted

Inspiration

When flights get delayed or cancelled, it creates a domino effect: passengers need rebooking, gate assignments change, and most critically, crew scheduling falls apart. Flight attendants and pilots have strict duty time limits (FAA regulations), and a single delay can cascade into crew timeouts, forcing additional cancellations. Currently, crew schedulers manually call hundreds of crew members to reassign flights, often taking 2-3 hours per major disruption.

Overview

Rerouted implements a parallel tabu search-based optimization engine for airline crew recovery under operational disruptions (e.g., flight delays). The system explores multiple independent optimization scenarios in parallel to quickly identify high-quality crew reassignment plans that minimize disruption impact while respecting FAA regulations, certification, location, and operational constraints.
Frontend features: [add frontend features]

Tech Stack

Frontend

Next.js 14 with App Router for server-side rendering
React for interactive UI components
Tailwind CSS for modern, responsive design
React Native (Expo) for cross-platform mobile development
Expo Notifications for push notification handling
React Navigation for screen routing
Shadcn for custom UI components

Backend

Node.js + Express for API server
PostgreSQL for crew data, flight records, and assignment tracking
American Airlines Flight Engine API for flight data integration
Python for logic engine and the optimization algorithm

Key Features

Parallel Tabu Search Algorithm

CPU-Efficient Execution
Realistic Airline Operations Modeling
- Uses flight data provided by AA Flight Engine
- Enhanced with crew, plane, pairing info (synthetically generated)
- Aircraft rotations in fleets
- Crew distribution
- Multi-leg pairings for crew members