Prediction Model Development for Sanitary Sewer Pipes’ Condition Assessment Using Logistic Regression and Neural Networks: Fid-Bau Portal

Prediction Model Development for Sanitary Sewer Pipes’ Condition Assessment Using Logistic Regression and Neural Networks

Atambo, Daniel O. / Najafi, Mohammad / Kaushal, Vinayak

Abstract

Sanitary sewer pipes’ infrastructure system in good condition is essential in providing safe conveyance of the wastewater from homes, businesses, and industries to the wastewater treatment plants. Most of the water utilities have aged sanitary sewer pipes. Water utilities inspect sewer pipes to decide which segments of the sanitary sewer pipes need rehabilitation or replacement. This condition assessment process is costly and necessitates developing a model that predicts the condition rating of sanitary sewer pipes. The main objective of this study was to develop a model to predict sanitary sewer pipes’ condition rating using inspection and condition assessment data. Sanitary sewer pipes were categorized based on various factors: physical, environmental, and operational. Physical factors included pipe material, diameter, age, slope, and depth. While the surface condition, soil type, corrosivity concrete, corrosivity steel, and pH were categorized under environmental factors, sanitary sewer pipe flow was deemed an operational factor. The environmental factors were used as independent variables and the condition rating score of the sanitary sewer pipes as a dependent variable in the model development using the city of Dallas’ data. The logistic model was built using 80% of randomly selected data and validated using the remaining 20% of data. The NNs model was trained, validated, and tested, and performed better than the logistic regression model. This study helps to predict sanitary sewer pipes’ condition rating that enables policymakers and sanitary sewer utility managers to prioritize the sanitary sewer pipes to be rehabilitated and/or replaced.