Journal of Infrastructure Systems

This research introduces the use of approximate dynamic programming to overcome a variety of limitations of distinct infrastructure management problem formulations. The form, as well as the parameters, of a model specifying the long-term costs associated with alternate infrastructure maintenance policies are learned via simulation. The introduced methodology makes it possible to manage large heterogeneous networks of facilities related by budgetary restrictions and resource constraints as well as by dependencies in maintenance costs or deterioration. In addition, the methodology is particularly well suited to consideration of multiple types of infrastructure condition data at the same time, including continuous-valued data and relevant historical data. Introduced techniques will prove valuable when high-quality deterioration and cost estimation models are available but are ill suited for use in a Markov decision problem framework. Computational studies show that the introduced approach is able to find an optimal solution to a relatively simple infrastructure management problem, and is able to find increasingly good solutions to a more complex problem.

Pipelines account for the lion’s share of the lifetime costs of any typical urban wastewater collection. While the deterioration of pipeline is the obvious “frontline” reason to adopt asset management practices, a more overarching driver is the need to gravitate toward sustainable development and set short- and long-term objectives encompassing economic, social, and environmental goals. Optimizing the expenses and providing the required levels of service to customers are the twin targets of wastewater utilities. Oslo Vann og Avløpsetaten—the water/wastewater utility in the Norwegian capital, considered as case study in this paper, is sprucing up its data bank, rightly realizing that a very effective information management system is the spine of asset management.

The evaluation of the effect of maintenance and rehabilitation (M&R) activities plays a critical role in pavement management systems. A quantification of the treatment effect usually resorts to simple statistical regression analysis. However, these analyses often suffer from two major problems: (1) sample selectivity bias due to the nonrandom nature of treatment applied to pavement sections and (2) inaccurate or erroneous representation of the treatment effect due to the difficulty in determining the timing between M&R activity and performance inspection. In this paper, an econometric technique, average treatment effect (ATE) modeling, is applied to address these concerns. In addition to its capability in addressing the two aforementioned problems, ATE modeling includes other desirable properties: (1) for any section individual, the treatment effect can be estimated regardless of whether or not it received treatment and (2) the treatment effect for specific sections of interest can be evaluated. A case study with focus on highway pavement overlay treatment is presented. The treatment effect on varying flexible pavement structure is investigated. A structured econometric ATE model is constructed and estimated, and a traditional regression model is presented for comparison. The comparison between the two models suggests that the ATE model is more accurate and informative in addressing the pavement maintenance treatment effect. The results can lead to profound implications for M&R decision-making and optimal budget allocation in pavement design and management.

Despite billions of dollars of annual exposure from claims and litigation related to construction-induced damage, there are no quantitatively based, agreed upon standards or procedures as to what constitutes due diligence with respect to a preconstruction, condition assessment. Similarly, the relative accuracy, reliability, and costs for various inspection approaches are not well established. This paper compares the relative performance capabilities of crack detection by sidewalk-based manual inspection with digital photography, terrestrial Light Detection and Ranging (LiDAR), and elevated manual inspections based on two brick and two concrete buildings (8.2–14.3 m high) in Dublin, Ireland. Results showed that nonmanual methods tended to overpredict crack widths by at least 5 mm and underestimate crack lengths by one-half. Digital photography, however, detected the shortest cracks (as short as 17 mm) and had no significant decline in accuracy beyond 12 m high, which has the added benefit of generating a permanent objective record. The terrestrial LiDAR proved neither particularly accurate nor cost-effective at the selected point density of less than $2\mathrm{mm}\times 2\mathrm{mm}$ . Finally, operator-based reliability problems emerged with all methods with discrepancies of at least 11%. Overall, digital photography taken and archived, but not analyzed, was the most cost-effective, accurate, and reliable approach.

In the aftermath of large-scale natural disasters, emergency management organizations are expected to provide safe temporary housing for a large number of displaced families and to ensure that these housing arrangements are not located in hazardous areas. Potential postdisaster hazards can take many forms such as earthquake aftershocks, landslides, postearthquake soil liquefaction, flooding, hazardous material releases, etc. This paper presents the development of a multiobjective optimization methodology to support decision-makers in emergency management organizations in optimizing postdisaster temporary housing arrangements. The developed methodology incorporates (1) a safety model to measure and quantify temporary housing safety in the presence of multiple potential postdisaster hazards; (2) a cost model to minimize total public expenditures on temporary housing; and (3) a multiobjective optimization model to simultaneously maximize temporary housing safety and minimize public expenditures on temporary housing. An application to a large region is presented to illustrate the use of the models and demonstrate their capabilities in optimizing postdisaster temporary housing arrangements.

As the need to better manage scarce water resources and water distribution systems increases, the problem of efficient routing of piping networks is gaining importance within the framework of an overall strategy for improving the networks’ efficiency and resilience to undesired emphoperational changes. The paper presents a methodology for optimizing flow routing in pipe networks by imitating the natural selection processes used by real-life ants in search of the shortest path to a food source. The method, known as ant colony optimization (ACO), is a population-based, artificial multiagent, general-search technique for the solution of combinatorial problems with its analogical roots based on the behavior of real-ant colonies. ACO’s mathematical background is outlined and a suggested possible implementation strategy is described for identifying “shortest paths” in water pipe networks. Such shortest paths could be not only the minimum pipe lengths between nodes of interest, but also the minimum number of valve operations required to keep a flow path active, the minimum number of customers affected during a flow reroute either because of planned (maintenance) or unplanned (water leak) conditions, and the minimum pressure drop along a path during adverse conditions. The ACO methodology should be of interest to both researchers and practitioners since it provides an alternative method to routing optimizations, with a wide range of applications. A case study of a specific urban water distribution network is also described for the proposed ACO virtual multiagent approach.

The use of titanium dioxide $\left({\text{TiO}}_2\right)$ ultrafine particulates as coating for concrete pavement have received considerable attention in recent years as these coatings can trap and absorb organic and inorganic air pollutants by a photocatalytic process. Despite these promising benefits, the promotion of ${\text{TiO}}_2$ coatings based on a single factor such as air quality does not provide a complete evaluation of this technology and may omit critical environmental factors that should be considered in the selection of sustainable materials. The objective of this study was to determine the life-cycle assessment of ${\text{TiO}}_2$ coating technology based on a hybrid life-cycle methodology. To achieve this objective, a life-cycle inventory that quantifies the energy, abiotic raw material inputs, and emission of ${\text{TiO}}_2$ coatings from cradle to grave were developed. Based on this inventory, life-cycle impact assessment of ${\text{TiO}}_2$ coatings for concrete pavement was conducted using the BEES impact assessment model. The use of titanium dioxide coating reduces the concentration of harmful pollutants by a photocatalytic process, and therefore, reduces the environmental impacts in four main categories: acidification, eutrophication, criteria air pollutants, and smog formation. However, during the production phases and due to the consumption of fossil energy, titanium dioxide will cause an increase in global warming, fossil fuel depletion, water intake, ozone depletion, and impacts on human health. Based on the overall environmental performance of this product, life-cycle assessment shows that titanium dioxide coating has an overall beneficial effect on the environment.

Storm-water pipe networks in Australia are designed to convey water from rainfall and surface runoff. They do not transport sewerage. Their structural deterioration is progressive with aging and will eventually cause pipe collapse with consequences of service interruption. Predicting structural condition of pipes provides vital information for asset management to prevent unexpected failures and to extend service life. This study focused on predicting the structural condition of storm-water pipes with two objectives. The first objective is the prediction of structural condition changes of the whole network of storm-water pipes by a Markov model at different times during their service life. This information can be used for planning annual budget and estimating the useful life of pipe assets. The second objective is the prediction of structural condition of any particular pipe by a neural network model. This knowledge is valuable in identifying pipes that are in poor condition for repair actions. A case study with closed circuit television inspection snapshot data was used to demonstrate the applicability of these two models.