Carbon cost calculator

final output

Inspiration

In today’s world, climate change and environmental sustainability have become global concerns. One major contributor to environmental damage is carbon emissions from vehicles. Most people are unaware of how much their vehicle contributes to carbon pollution, and there's often no simple way to measure it in real-time.

That’s what inspired us to build the Carbon Cost Calculator — a tool that uses Machine Learning to estimate the CO₂ emissions of a vehicle based on parameters like engine size, fuel consumption, and vehicle class.

What it does

The Carbon Cost Calculator is a web-based tool that uses Machine Learning to estimate the carbon emissions of a vehicle based on its specifications. It’s built using Streamlit for the front end and trained using real-world vehicle data.

How we built it

We developed the Carbon Cost Calculator using a combination of Machine Learning, Python, and Streamlit to create a user-friendly and intelligent tool that predicts carbon emissions based on vehicle data.

Challenges we ran into

Missing or Inconsistent Data Mismatched Column Names Model Overfitting Streamlit App Styling Issues Integrating MySQL to store predictions caused connection issues and insert errors.

Accomplishments that we're proud of

Successfully combined machine learning, web development, and database integration into one cohesive too Developed and trained a Linear Regression model that accurately predicts CO₂ emissions based on vehicle specifications. Created a clean and responsive web interface using Streamlit that anyone can use — even without technical knowledge. Gained experience in full-stack development — from preprocessing data and training a model, to deploying an app with a frontend and database.

What we learned

Learned how to train, test, and evaluate a regression model using real-world data. Realized that cleaning and preparing data is often more time-consuming than model training. Learned techniques like label encoding, handling missing values, and feature scaling. Understood how to connect different technologies: Python, Streamlit, MySQL, and Pickle. Learned how to build a project that works seamlessly from user input to database logging. Gained experience using Streamlit to create a simple, interactive, and visually appealing UI for non-technical users. Faced issues like mismatched column names, styling problems, and SQL connection errors — and learned how to solve them systematically. Learned more about carbon emissions, fuel consumption, and the role of transportation in climate change. Understood how tech can be used to raise awareness and inspire behavioral change.

What's next for Carbon cost calculator

The Carbon Cost Calculator is just the beginning. There are many exciting directions we plan to take this project to increase its accuracy, usability, and impact Add Climate Impact Insights Improve prediction accuracy by considering climate conditions and traffic patterns, which influence fuel consumption and emissions. Allow users to enter start and end points to estimate carbon emissions per trip, helping them choose more eco-friendly travel routes. Build an Android/iOS version so users can calculate their carbon footprint on the go and log data during actual trips. Experiment with more powerful models like Random Forest, XGBoost, or Neural Networks for better prediction accuracy.