| description abstract | Wastewater pipes age and deteriorate over time, and utilities face great challenges in operating and maintaining their wastewater pipe infrastructure. Thus, systematic assessment and evaluation programs are required for efficient monitoring of the conditions and performance of wastewater pipes. This paper presents a rating system for use in evaluating the condition of wastewater pipes that can assist utilities in planning, prioritization, maintenance, and repair/rehabilitation/replacement decisions. A condition-rating system also can be an extremely useful tool in benchmarking pipe conditions, which allows the comparison of pipe segments within the system with each other and with similar segments in other systems. Currently, many utilities use the National Association of Sewer Service Companies’ (NASSCO’s) pipeline assessment and certification program (PACP) or have developed their own, in-house condition-rating system based on PACP. Most of the condition-rating systems currently in use are based on structural and operational defects of pipes. The proposed methodology considers defects identified from inspections, e.g., cracks, holes, and corrosion, and other parameters that affect the conditions and performance of wastewater pipes, such as soil characteristics, loading, and flow velocity. The proposed performance-rating system evaluates each parameter and combines them mathematically through a weighted summation and a fuzzy inference system that reflects the importance of the various factors. The framework provides a noticeable improvement from the conventional practice of using solely inspection data as a means to evaluate the wastewater pipe infrastructure system, because the model accounts for inspection data and other parameters that influence wastewater pipe performance. The fuzzy logic was good for approximate reasoning and incomplete information. The fuzzy inference model was found to have many advantages over the weighted factor model. The fuzzy inference model provided more sensitive results, whereas the weighted factor model provided more static results since the weights assigned to each parameter were fixed and distributed throughout all parameters in the model. In addition, the fuzzy inference model accounted for the combination effect of dependent parameters. | |
