Condition Assessment of Reinforced Concrete Sanitary Sewer Pipelines Using Probabilistic Methods and Advanced Inspection Tools

Abstract

Large sanitary sewer pipelines (SSPs) are the backbone of modern infrastructure. They carry wastewater from smaller lines to treatment plans; therefore, they are critical for public health and safety. According to the ASCE report card published in 2021, the condition rate of sanitary sewage networks in the United States is D+ (i.e., poor) (ASCE 2021). Therefore, frequent inspections of SSPs are crucial for performing a proper life cycle management strategy. New inspection technologies such as LiDAR have recently been employed for rapid condition assessment of SSPs. Because of the precision of LiDAR inspection data, it can not only measure hydraulic properties but also, quantify the erosion of concrete walls in SSPs. Because of the limited available inspection data for this large aging inventory, probabilistic approaches need to be implemented into these limited inspection data to create an effective condition assessment for SSPs. Meanwhile, Reinforced concrete SSPs (RCSSPs) are commonly used for sewer mains. In this study, an automated framework for condition assessment of RCSSPs is proposed using LiDAR inspection data and probabilistic approaches. The framework includes the procedure for filtering and alignment of the raw 3D point cloud of data, which represents the pipe's inner geometry coordinates. Then hydraulic properties (such as hydraulic radius, wetted perimeter, volumetric flow rate, and velocity of flow), ovality, and concrete erosion rate from the 2D cross-sections along the pipe length. From these outputs, probabilistic approaches will be implemented such as finding the best-fit probability density function (PDF) corresponding to the deteriorated inner concrete wall geometry of RCSSPs, and probabilistically estimating their residual service life (RSL); in addition, the Bayesian network is used to predict the erosion rates using different priors from the literature and the likelihood function which is the calculated erosion rate from the inspected SSPs at the age of 28. The results will be compared in the form of the predicted remaining service life (RSL). Lastly, statistical approaches such as single/multi-variable regression, and polynomial regression are applied to the outputs to estimate the mean concrete loss and identify underlying patterns in the data. The overall objective of this study is to develop an innovative, automated, and rational framework for condition assessment of transportation infrastructure more specifically for RC sanitary pipelines.