Prioritizing Sewer Line Inspection with an Expert System

Abstract

This paper describes the development of an expert system that prioritizes sewer inspection. The expert system also identifies potential risks and their consequences, provides an appropriate inspection test method, and alerts the user when more information is needed to make an accurate determination. Prioritizing sewer inspection identifies the critical reaches of the system, reduces the number of inspections, and as a result, reduces both emergency repair and preventative costs. The belief network logic of the expert system is based on information to be provided by the sewer utilities of ten medium-sized cities. The utilities will provide expert input on the probabilities of occurrence and relational dependencies associated with sewer line defects, condition assessment, and maintenance and rehabilitation procedures. This information helps assign the probability weights and relationships of the system's knowledge base. Feedback on prototype evaluations from the utilities will aid the validation process of the expert system development. IntroductionMany cities are being serviced currently by non-inspected and deteriorating sewer infrastructure. Factors, such as weak materials, poor construction practices, and inadequate management, have made sewers less resilient to the natural aging process (Iseley, 1997). Examples of these factors are softened or absent mortar in brick lined sewers, improper trench compaction and materials, poor tunneling, infiltration and inflow, hydraulic overloads, and neglect. Deteriorating sewers leave communities vulnerable to unexpected catastrophic failures that disrupt not only sewer service but also above-ground activities. Failures are difficult to avoid if the cities are unaware of their pipes' condition.The need to prioritize sewer inspection is a result of the uncertainty of the condition of buried infrastructure and the potential for costly and catastrophic system failures. By prioritizing sewer inspection, timely information can be obtained about the most critical reaches. Scheduling sewer inspection and the subsequent maintenance/rehabilitation process is a part of a utility's larger management plan.The expert system described in this paper will not replace existing management plans, but enhance ongoing planning efforts. Many infrastructure management plans are focused on prioritizing rehabilitation. However, prioritizing rehabilitation requires a detailed inspection of a sewer system's lines. Therefore, prioritizing inspections is

Application of Gamma Process for Deterioration Prediction of Buildings from Discrete Condition Data

“…Among the AI based techniques, Case-Based Reasoning (CBR), fuzzy set theory and Neural Networks (NNs) are used for modeling the deterioration of infrastructure facilities. CBR is used for sewer [8]. Fuzzy set theory is used in buried pipes [9].…”

A Spatial Statistic Based Risk Assessment Approach to Prioritize the Pipeline Inspection of the Pipeline Network

“…The most popular version of objective probability is frequentist probability, and the most popular version of subjective probability is Bayesian probability [15][16][17]. Thus, risk-based inspection stands two lines: frequentist probability based methods [18][19][20][21] and Bayesian probability based methods [12,14,[22][23][24][25][26][27][28][29]. The former obtains so-called stochastic or aleatory uncertainty, which is estimated by the application of probability models based on a large amount of relevant data.…”