Commit to the Middle

Inspiration

With teams spread across continents, every meeting location choice creates winners and losers: some offices always get the short flights while others rack up the long-haul miles and carbon emissions. We wanted to build a tool that makes the invisible trade-offs visible. How much CO₂ are we really emitting? Which offices are bearing the travel burden? Is there a fairer middle ground?

What it does

Commit to the Middle is a meeting location optimizer that helps distributed teams find the sweet spot between sustainability and fairness. Given a list of attendee locations and meeting requirements, our tool:

Analyzes real flight routes from OAG's comprehensive schedule data (214,000+ daily flights) Calculates actual CO₂ emissions based on aircraft type, distance, and passenger load factors Computes round-trip travel times including connections and layovers Measures fairness using statistical variance to ensure no office is always stuck with the worst journey Compares candidate cities across multiple optimization criteria: eco-focused, time-focused, fairness-focused, or balanced

The output provides valuable insights: top recommended meeting locations, total CO₂ footprint, travel time breakdowns per office, and a fairness score all with full transparency into the trade-offs.

How we built it

We divided the project into parallel workstreams to maximize our 24-hour hackathon time: Data Pipeline (Team Member 1)

Built CSV streaming engine to query OAG's massive schedule database efficiently
Created emissions lookup system matching flights to CO₂ data (328,872 emission records)
Implemented city-to-airport mapping for 60+ global cities
Developed flight filtering and sorting algorithms (fastest, eco-friendly, shortest, etc.)

Optimization Engine (Team Member 2 )

Designed metrics calculation system for CO₂, travel times, and fairness
Implemented location comparison algorithm with multiple optimization modes
Built event scheduling logic (determining when meetings can start based on arrivals)
Created weighted scoring system for balancing competing objectives

Integration & Testing (Team Member 3)

Developed JSON input/output interface for automated testing
Created main orchestration pipeline connecting all components
Built validation and error handling systems
Tested across diverse scenarios (same-region vs intercontinental, balanced vs unbalanced attendance)

Visualization (Team Member 4)

Focused on building front front-end and UI
Made graphics and design
Connected front-end with back-end
APIs and

Key Technical Decisions: Round-trip calculations: We model the full journey, not just outbound flights, because sustainability means counting all emissions Fairness metric: Standard deviation of travel times ensures equitable burden distribution Modular architecture: Clean interfaces between components enabled true parallel development

Challenges we ran into

The Round-Trip Bug Initially, our calculations only counted one-way flights. This made every solution appear 50% better than reality! We caught this during integration testing and had to refactor the metrics engine.
The Zero-Hours Mystery Cities were showing 0.0 travel hours when they shouldn't. We were treating "no flights found" as "already local." Required careful distinction between missing data vs. zero-cost scenarios.
Performance Bottleneck Loading 328K emission records for every city comparison was crushing performance. Solution: implement caching and optimize CSV streaming.
Front-End Integration Connecting our Python backend with the visualization layer required careful API design and data format alignment.
Exhaustion Working through the night, we hit the wall around 4 AM. Coffee and team motivation got us through the final debugging session.

Accomplishments that we're proud of

Working end-to-end solution that handles real-world data and produces actionable recommendations Client-first thinking: We focused on making results interpretable and useful for decision-makers, not just technically correct Algorithmic innovation: Created our own weighted scoring system that elegantly balances multiple competing objectives Data integrity: Properly accounting for round-trips, return flights, and all hidden costs—no greenwashing here! Fairness as a first-class metric: While most solutions optimize for total CO₂ or total time, we put equity front and center Scalability: Works for any number of cities and attendees, from 2-person meetings to company-wide gatherings

What we learned

Technical Skills: Large-scale data processing with CSV streaming Multi-objective optimization algorithms API design for parallel development The importance of test-driven development (those bugs taught us!)

Domain Knowledge: Flight emissions factors vary dramatically by aircraft type and load factor Geography matters: a "central" location isn't always the fairest Small changes in meeting dates can unlock better flight options

Soft Skills: Breaking complex problems into parallelizable components Communicating trade-offs to non-technical stakeholders Debugging under pressure (and sleep deprivation!) The power of code reviews, catching the round-trip bug saved our demo

What's next for Commit to the Middle

Immediate Enhancements Last-mile transportation: Add taxi/train CO₂ from airport to actual office locations Cost optimization: Factor in flight prices alongside emissions and time Multi-day meetings: Optimize hotel stays and local transportation

Advanced Features Alternative transport: Suggest trains for shorter routes (<500km) with lower emissions Historical analysis: "How much CO₂ did we emit last year? What if we'd optimized?" Dynamic scheduling: Find the best meeting dates within a window, not just best location Real-time pricing: Integrate with flight APIs for cost-aware recommendations