No abstract

Drinking water and natural gas are essential goods provided by municipal distribution networks. In order for there to be a reliable provision of such commodities, infrastructure managers must regularly dispatch vehicles to perform tasks such as valve and hydrant inspections and meter replacements. As public concerns about sustainability grow, managers must look for ways to reduce carbon emissions. This, however, should be done without jeopardizing service. One conceivable approach is to train staff to perform additional maintenance tasks so that combined work packages can be carried out. This would require resources to be spent in different areas, such as supplemental worker training and organizational restructuring of the utility. Infrastructure managers must thus weigh the expected efficiency gains against the associated costs. The infrastructure asset management process combined with digitalization are powerful tools to assess such questions. Unfortunately, digitalization in the infrastructure sector lags behind other sectors of the economy. To bridge this gap, real-world examples are needed to further spur adoption. This paper addresses this need with a methodology and case study for infrastructure managers of water and gas networks. Specifically, this paper presents a methodology to quantify the resource requirements of operational maintenance programs for a large municipality. As utilities plan maintenance routes differently, four algorithms are used to model the resource requirements of the status quo. The effect of prioritization as well as frequency of inspections/replacements is considered.

Cities rely heavily on the services provided by water distribution networks. These networks are large and complex, consisting of thousands of kilometres of buried pipes and dozens of facilities where water is treated, pumped and stored. Infrastructure managers are entrusted with the planning and execution of interventions on these assets to ensure that the provided service exceeds the minimum levels mandated by stakeholders at all times. This is a difficult task due to the spatial extent of these networks, shrinking budgets and the complexity of coordinating with multiple stakeholders. Previously, Kerwin and Adey presented an approach to address these concerns, using a small example network, leaving open the question of how this approach would work on a large real-world network. This paper fills this gap by discussing the simplifications needed to apply the methodology to a larger network and demonstrating its advantages with three applications: (a) estimating the budget requirements needed to implement various intervention strategies, (b) communicating project-level trade-offs of different intervention strategies and budget scenarios and (c) investigating how the intervention-planning activities of other networks could affect these estimates. The methodology is demonstrated on a large distribution network consisting of 14 pressure zones for a 5-year period.

Buried pipes comprise a significant portion of assets of a water utility. With time, these pipes inevitably fail. Failure prediction enables infrastructure managers to estimate long-term failure trends for budgetary planning purposes and identify critical pipes for preventive intervention planning. For short-term prioritization, machine learning based algorithms appear to have superior predictive performance compared to traditional survival analysis based models. These models are typically stratified by material resulting in the exclusion of newer pipe materials such as polyethylene and corrosion-protected ductile iron, despite their prevalence in modern networks. In this paper, an application of an existing methodology is presented to estimate time to next failure using artificial neural networks (ANNs). The novelties of the approach are 1) including material as an input parameter instead of training several material-specialized models and, 2) addressing rightcensored data by combining soft and hard deterioration data. The model is intended for use in short-term prioritization.

No abstract