Comprehensive Feature Analysis for Sewer Deterioration Modeling

Timely maintenance of sewers is essential to preventing reduced functionality and breakdown the systems. Due high costs associated with inspecting a sewer system, substantial research has focused on deterioration modeling identification most useful features. However, there lack consensus in findings. This study investigates how feature importance depends definition bad pipes changes between utilities similar data bases. A dataset containing 318,457 from 35 condition state (CS) ranging one four was used. The cleaned, backward step analysis (BSA) applied two ways binarizing CS. Additionally, BSA for each utility ≥100 CS four. results showed that selective performance changed order which features contributed most. In case, either year construction, age, groundwater, rehabilitation, or dimension important feature. On average 6.5 utility-specific models. sensitive inspection strategy, size dataset, interdependency