List All Projects   List Graduate Projects   List Senior Design Projects   List REU Projects   Other Undergraduate Projects

Project # 71
A Supervised Machine Learning-Based Cost Estimation and Prediction Model for Wastewater System Rehabilitation Using Excavation and Trenchless Technologies

Team Lead: Alexis Reyes
Other Team Members:
Faculty Advisor(s): Vinayak Kaushal, Ph.D., P.E.
Department: CE
Project Type: REU
Title: A Supervised Machine Learning-Based Cost Estimation and Prediction Model for Wastewater System Rehabilitation Using Excavation and Trenchless Technologies
Abstract: Wastewater pipelines, like other critical infrastructure assets, are vulnerable to structural failures, blockages, and overflows. With the U.S. wastewater infrastructure currently graded D-, and the overall national infrastructure rated C-, there is an urgent need for cost-effective rehabilitation strategies. Trenchless technologies, particularly cured-in-place pipe (CIPP), have emerged as innovative alternatives to traditional excavation technologies (ETs), offering potential economic and environmental advantages. However, a comprehensive cost estimation and prediction studies comparing these approaches remains underexplored. This study introduces a supervised machine learning-based model for cost estimation and prediction in wastewater system rehabilitation, focusing on both ETs and trenchless CIPP technologies. The model integrates construction, environmental, and social cost components to evaluate the net present value of each method over its service life. Using data from polyvinyl chloride (PVC) and ductile iron (DI) sewer pipes ranging from 8 to 14 inches in diameter, the model predicts cost outcomes and sustainability impacts. Results indicate that while PVC offers lower initial material costs, DI pipes demonstrate superior long-term cost-effectiveness and environmental sustainability. The proposed intelligent model provides engineers and decision-makers with a data-driven tool to assess and compare rehabilitation strategies, enabling more informed infrastructure investment decisions.
Poster: View Poster #71 (PDF)
Judges Scoring: Submit Evaluation (judges only)