Structural Condition Assessment of Sewer Pipelines

Abstract

The need of immediate supportive measures for sustainability of municipal infrastructures calls for better understanding of the behavior of various infrastructure network systems and their components. This paper presents a study which uses artificial neural networks to investigate the importance and influence of certain characteristics of sewer pipes upon their structural performance, expressed in terms of condition rating. In this study, back propagation and probabilistic neural network NN models were developed and validated. The data used in the development of these models were provided by the municipality of Pierrefonds, Quebec. It comprised of parameters related to sewer pipelines, pipe diameter, buried depth/cover, bedding material, pipe material, pipeline length, age, and closed circuit television CCTV based structural condition rating. The first six parameters are the independent variables of the models whereas CCTV based condition rating for these pipes is the dependent variable i.e., the output of the models. The developed NN models were used to rank the parameters, in order of their importance/influence on pipe condition. It was found that, among the studied parameters, material attributes have highest influence on pipe structural condition, respectively, followed by the geometric and physical attribute group. Sensitivity analysis was then performed to simulate the structural condition of a pipe at a range of values of each input parameters. Results of sensitivity analysis describe the nature and degree of the influence of each parameter on pipe structural condition. The developed models are expected to benefit academics and practitioners municipal engineers, consultants, and contractors to prioritize inspection and rehabilitation plans for existing sewer mains. DOI: 10.1061/ASCECF.1943-5509.0000081 CE Database subject headings: Neural networks; Assessment; Parameters; Sensitivity analysis; Sewer pipers; Rehabilitation. Author keywords: Neural networks; Condition assessment; Parameters; Sensitivity analysis; Sewer mains.