Infrastructure Condition Assessment, Deterioration Modeling, and Maintenance Decision-Making: New Contributions for Improved Management 

Published online ahead of print on March 4, 2008This special issue is the second of two, reflecting recent advances in infrastructure condition assessment, deterioration modeling, and optimal maintenance, repair, and reconstruction. The first special issue (September 2006) featured methodological advances that are taking place in light of practical considerations. In this second issue, the need to improve upon the management of infrastructure facilities and systems by explicitly addressing practical matters is again a recurring theme. In addition, the papers in this issue remind us that much technological and methodological research is also needed to help the practicing community effectively and efficiently address its pressing demands.

Cascante et al. develop a fuzzy-logic based methodology for assessing the condition of hydraulic channel sidewalls, based on nondestructive testing and expert knowledge. Collecting data with nondestructive methods reduces cost, maintains serviceability, and preserves historically significant structures. Incorporating expert knowledge is often a cost-efficient means of assessing infrastructure condition. Therefore, capturing this knowledge in a rigorous, consistent, and valid manner has value to both the research and practicing communities. The authors support the utility of their method by showing the reasonableness of its sensitivity to the maintenance action applied in a realistic setting.

Prozzi and Hong develop a multivariate deterioration model that addresses multiple pavement condition measures relevant to the management of infrastructure facilities. Their rigorous statistical approach simultaneously models the deterioration mechanisms corresponding to the different measures, while capturing the correlation among these measures. The authors also present an extensive case study, exploiting data collected from in-service facilities to demonstrate the importance of considering the multivariate nature of deterioration.

Numerous pavement deterioration models have been developed and estimated using field data over the years. These models attempt to capture different characteristics of the data, represent deterioration behavior through alternative model structures, and rely on various statistical techniques in estimating the corresponding model parameters. While such developments have been compared qualitatively, Chu and Durango-Cohen conduct a rigorous quantitative comparison of a subset of models estimated with a widely used empirical data set. Their quantitative comparison can help researchers determine the most important modeling issues to address in future studies. It can also help practitioners compare more rigorously the advantages and disadvantages of using each of the models for pavement condition forecasting.

Most maintenance decision-making approaches employ a discrete representation of condition. However, approaches to decision-making based on a continuous representation have recently been receiving serious attention. Advances in continuous condition based decision-making are of interest to the infrastructure management community, since infrastructure damage variables are, for the most part, inherently continuous in nature. Rapid advances in automated inspection techniques are making these damage variables more easily measurable and readily available, and practical benefits from considering this more natural representation of condition are increasingly possible. Jido et al. advance the ability to arrive at meaningful maintenance decisions, based on a continuous condition representation, by considering joint optimization of inspection and maintenance timing under forecasting uncertainty and demonstrating an ability to develop solutions to their rigorous formulation. These advances should foster further research in formulating, solving, and implementing infrastructure management methods using continuous representations of important condition variables.

Budgetary matters are of great concern to infrastructure providers and managers, and, in some sense, characterize the essence of the infrastructure management problem. Budget related considerations are generally addressed in the infrastructure management literature by including one or more budget constraints to system-wide optimization formulations. Kobayashi et al. offer a different approach by explicitly incorporating budgetary accounting principles within pavement management systems. Their approach allows infrastructure managers to justify maintenance budget requests based on accepted cost accounting information and, in effect, marries pavement deterioration with asset depreciation. The authors present an application example in a realistic context that demonstrates the feasibility and value of this new framework.

In summary, the developments reflected in this special issue make it clear that important contributions are still needed in addressing the various challenges posed by the management of complex and costly infrastructure systems. Effectively addressing these challenges will require both operationalizing previous advances, and developing new methods and approaches that further focus on the geographic extensiveness, long life spans, the uncertain nature of infrastructure systems.