Closure to “Integrated Decision Support System for Optimal Renewal Planning of Sewer Networks” by M. R. Halfawy, L. Dridi, and S. Baker

Abstract

The authors of this paper discussed a number of issues in wastewater infrastructure management, including condition rating, deterioration modeling, risk assessment, asset prioritization, and optimization of rehabilitation strategies within budgetary constraints. While their effort is certainly laudable, I want to offer some comments about the deterioration modeling as presented in the “Example System Use” on p. 368, as well as discussed elsewhere in the paper. The authors recognize the importance of quality and quantity of condition assessment data to develop reliable deterioration models for sewer networks, but they fall short on the importance of using appropriate and correct model development and validation procedures. For example, the paper suggests and applies ordinary regression model s for deterioration of wastewater pipelines. However, it may not be advisable, and be “theoretically inappropriate” Lynch 2007 to use ordinary regression model s to predict discrete and ordered outcomes for the following reasons: 1. The deterioration may be intrinsically continuous, but it is usually measured in discrete and ordinal i.e., noncontinuous ways, in terms of, for example, structural internal condition grades ICGs or structural performance grades SPGs for wastewater pipelines. As noted by the authors, the Water Research Center’s Sewerage Rehabilitation Manual suggests