A Fuzzy Expert System for Prioritizing Rehabilitation of Sewer Networks

Abstract

Abstract: Rehabilitation of sewer networks is a huge and very costly global problem that has often been treated on a crisis-based approach. The development of a rehabilitation program requires models and tools for assessing the condition and performance of sewers. The original contribution of this study is the development of a ranking scheme for sewer rehabilitation priorities. A fuzzy expert system was applied with inputs from a combined assessment of hydraulic, structural performance and potential failure consequences. The fuzzy structural system computes the global structural performance index for each pipe using internal condition, surrounding condition, and site vulnerability (SV) as inputs. The fuzzy hydraulic system uses hydraulic performance index (HPI), hydraulic performance impact, and SV to compute the global HPI. Finally, the fuzzy global system uses all these factors to compute the global performance index for each pipe. This methodology was successfully applied to the sewer system of the City of Laval in Canada. The results show how the fuzzy inference system may be used to establish rehabilitation priorities for each pipe section. The fuzzy expert system provides more realistic results than the intuitive approaches that use structural and hydraulic performance maximum and mean.