Document Type


Publication Date



Decision support for sewer rehabilitation programs is dependent on reliable information about the asset conditions. Inference about the condition development can be made by combining condition observations with deterioration forecasting models. GompitZ is a Non-Homogenous Markov Chain model, which enables waste water utilities to predict the sewer deterioration at a network segment level, based on closed-circuit TV (CCTV) inspection and normalized condition grading. The predictions can be used to assess investment needs for the sewer systems. Successful calibration and prediction is however dependent on the validity of the model assumptions, sufficient amounts of CCTV observations, and that the condition grading system used is capable of describing the actual sewer conditions. In this work, Monte Carlo simulations are used to assess the sensitivity of GompitZ calibration results with respect to available input data. The simulations are performed by starting with a full dataset of observations, and randomly selecting a subset from the full dataset. By repeating the calibration process many times for random subsets of equal size, one can assess the distribution of the model parameters and the uncertainty in the output from GompitZ. The sensitivity analysis is performed both on a real (Oslo municipality) and a virtual dataset. GompitZ uses a simplified Mixed Generalized Linear Regression technique, and the sensitivity results can be used to draw several conclusions about this calibration method and possible ways to improve it, as well as about the predictive capabilities of the model. Practical learnt lessons are expected, as for assessing: The inherent uncertainty in sewer conditions; i.e. at what point increased amounts of CCTV observations does not help to reduce the prediction uncertainty How the uncertainty in the predictions develops as the amount of data reduces The critical number of observations needed in order to achieve a successful calibration with GompitZ


Session R52, Sanitary Sewer Networks



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.