Date of Award

Spring 2001

Document Type


Degree Name

Doctor of Philosophy (PhD)


Civil Engineering

First Advisor

Robert A. McKim


Sewers represent a very substantial and crucial part of our underground infrastructure and a high level of investment is needed to maintain them. Sewers are generally out of sight and, as long as they continued to function reasonably well, they are also out of mind—that is, until any problems such as structural, hydraulic and functional arise. The criticality of any sewer condition is a combination of the risk of failure and the consequences of it. Therefore, a systematic approach is necessary to fully understand and evaluate the conditions of this underground infrastructure. The present state of the art does not allow the utility manager to determine which pipe distresses should be repaired immediately to stop further degradation, or which distresses are stable and could be left for a future repair.

The preliminary work in sewer prediction modeling reported in the author's Master's thesis (Kathula, 2000) has laid the groundwork for this research effort to develop a field-tested sewer-deterioration prediction model. In this research a detailed questionnaire was mailed to municipalities in the Gulf Coast region of the United States. The survey participants included municipal officials and experts from Alabama, Florida, Louisiana, Mississippi and Texas. These regional data were used to develop a regional sewer deterioration prediction model.

The sewer deterioration models were developed for the five common degradation states—cracks, open joints, displaced joints, corrosion and deformation for both clay and concrete sewers. This study used the basic principles of probability-based Markovian Chain methods to develop a distress transition matrix for sewer pipes based on the experience of municipal managers and sewer experts. The predictive model developed in this research has the ability to predict the service and structural distresses of clay and concrete pipelines for any given number of years, based on any initial condition of the pipeline. An effective and proven predictive model will allow the utility manager to apply life-cycle-costing economics to the utility.

The sewer deterioration prediction models developed for clay and concrete sewer pipes were verified by using the field data collected from municipalities using a concept called “risk ratio.” The risk ratios were calculated for various distress conditions at different time periods for the developed predictive model based on the expert opinion and were compared against the risk ratios calculated for validation data.