Pavement Condition Index Prediction Using ML

Problem Statement

Manual PCI assessment is resource-heavy, subjective and expensive for large road networks.

Objectives

• - Preprocess traffic, climate and distress datasets
• - Train ML models for PCI prediction
• - Automate segment classification into quality categories
• - Integrate outputs in GIS decision workflow

Methodology

The project follows a structured implementation approach that includes Data engineering with distress, traffic and climate features, Regression models: RF, GBM, XGBoost, LSTM, Classification models: SVM, DT, RF, and Evaluate with RMSE/MAE and classification metrics. These steps are executed iteratively to validate assumptions, improve performance, and ensure reliable delivery of the final solution.

Expected Outcome

By the end of this project, the team is expected to deliver Streamlit interface for non-technical users, and Automated PCI analytics for proactive maintenance. Together, these outcomes reflect both technical feasibility and practical value for demos, evaluation, and future scaling.

Future Scope

• - IoT sensor data integration
• - Satellite imagery and deep learning extensions