Haas Race Intelligence System (HRIS)

Desktop view of the Server. Displays real-time conditions: Lap 1, Rain, 23-lap tires, 97.5% fuel. Data collection in progress.
Python Flask engine running the server. Shows clean structure (model.py, strategy.py) and network readiness on 192.168.216.53:5000(can vary)
The HASS-F1-T3AM GitHub repo. All project code is here, including the core logic files (app.py, strategy.py, model.py).
Phone view - A strategy director isn't tied to a desktop; they're moving around the garage. They can see the phone flash a life-saving alert

Inspiration

Every Formula 1 race is a story of precision, data, and milliseconds. One wrong pit call can flip a victory into a loss. We were fascinated by how teams like Haas F1 constantly balance tire degradation, fuel load, and race pace — all in real time, under extreme pressure.

That curiosity sparked an idea: What if we could build a digital race strategist — one that doesn’t replace the engineers, but helps them make faster, smarter, data-driven decisions?

That’s how the Haas Race Intelligence System (HRIS) was born — a lightweight, intelligent decision engine designed for the pit wall.

What it does

The Haas Race Intelligence System acts like a virtual strategist sitting beside the pit crew.

It:

Monitors live race data — lap times, tire wear, weather, and fuel usage.

Creates a digital twin of the car to predict performance in different conditions.

Runs real-time simulations for pit stop vs. stay-out scenarios.

Suggests optimal race strategies — when to pit, what tire to use, and how it impacts total race time.

Displays everything in a clean, mobile-friendly web dashboard accessible from any device.

In short, it turns race chaos into clarity.

How we built it

Backend: Created a data simulation system to mimic live race telemetry using Python and Flask, sending live performance data streams.

Machine Learning: Trained a Linear Regression model (scikit-learn) to predict lap time deltas and tire degradation patterns, continuously adapting as laps progress.

Frontend: Built a lightweight dashboard using HTML, CSS, and JavaScript, optimized for both laptops and phones.

Infrastructure: Runs locally without heavy servers — relying on smart design and teamwork.

Challenges we ran into

Real-time simulation: Handling updates every few seconds without lag.

Model tuning: Balancing prediction accuracy with speed.

Cross-device testing: Ensuring consistent performance on laptops, tablets, and phones.

Networking: Configuring LAN access while avoiding firewall/port issues.

Accomplishments that we're proud of

Built a fully functional live race simulation server from scratch.

Integrated machine learning and analytics into an accessible web app.

Ensured cross-device usability (phones, tablets, PCs).

Transformed raw racing data into actionable insights.

Collaborated like a real F1 pit crew — fast, precise, and focused.

What we learned

The power of data-driven decision-making in motorsport.

How real-time systems differ from static ML models.

The importance of collaboration under pressure.

Balancing innovation with simplicity — making complex insights easily understandable.

What's next for Haas Race Intelligence System (HRIS)

Integrate real F1 telemetry datasets for authentic race conditions.

Expand from pit strategy to car performance optimization, predicting lap-time improvements from setup tweaks.

Deploy HRIS on the cloud for remote pit-wall access and multi-user collaboration.

Add a voice-enabled interface — engineers could ask “What’s the best lap to pit?” and get instant AI-driven answers.

Our goal is to evolve HRIS into a full-scale race intelligence platform supporting Haas F1 engineers in real-time — enhancing strategy, speed, and smart decision-making.

Built With

Submitted to

TrackShift Innovation Challenge

Created by

Backend Development & Model Implementation

I handled the core backend logic using Python and Flask.

Created the Flask routes to generate and send simulated race data (lap times, tire wear, weather impact) to the front-end.

Integrated NumPy and Scikit-learn for data simulation and predictive modeling, helping estimate optimal pit laps.

Ensured smooth API responses and real-time updates between the dashboard and the server.

Also tested and debugged Flask connectivity across devices on the same Wi-Fi network for multi-device use.
This backend layer became the “brain” of the system, connecting data, strategy, and visuals into one seamless experience.

Namratha Namratha
I focused on the research and implementation side of the project. My work included :

Studying real F1 race strategy systems — how teams like Haas, Mercedes, and Red Bull use telemetry data (lap times, tire degradation, weather changes, safety cars) to make instant pit-stop decisions.

Designing the architecture of the Flask backend and planning how data flows between the predictive model, the API, and the front-end dashboard.

Creating and organizing the GitHub repository, structuring folders for app.py, /templates, and /static to make the project easy to collaborate on.

Managing version control, commits, and pull requests to ensure smooth teamwork and a consistent codebase.

Outside academics, I’m genuinely passionate about Formula 1 — I follow races, strategy calls, and data analysis closely. That love for racing is what inspired me to build something that mimics the real pit-wall decision systems used in F1.
This personal interest helped me dive deeper into how data and strategy intersect in motorsport, and shaped the technical backbone of this project.

Prerit Kumar
I contributed to designing the dashboard interface using HTML, CSS, and JavaScript.
HTML:-Created a clean and minimal dashboard layout (index.html) to display live telemetry data like lap time, tire age, and weather.
CSS: Styled the interface with a dark racing theme (style.css) — added container design, colors, shadows, and typography for better readability.
JavaScript: Added auto-refresh functionality (script.js) to fetch and update race data from the Flask backend every 5 seconds, ensuring a real-time experience.
This front-end integration made the race strategy dashboard interactive and visually engaging.

URVI SINGH