Water distribution systems in the Department of Defense (DoD) are aging and require proactive management and replacement. In general however, a reactive approach for repair and replacement has been the norm across much of the DoD, operating under the out-of-sight out-of-mind mantra. The DoD has invested in other asset management procedures and software, such as BUILDER, to prioritize repair and investment into facilities. However, no such system exists for water distribution infrastructure. As such, this project plans to create a new tool to consider existing condition of assets, project asset deterioration, and model investment strategies. The tool will provide service, major command, and installation-level insights to aging water infrastructure to understand investment strategies and their impact on the overall system health. The tool will greatly advance the prioritization and investment strategies of the DoD in its aging water infrastructure to promote resilience and sustainment.

The technology will consist of an enhanced tool to facilitate asset deterioration forecasting over a categorical space. The tool will rely upon a probabilistic, multivariate Markov Chain probability transition model to simulate investment strategies that sustain or improve condition of water mains over time. Additionally, the tool will readily incorporate new data and update the projections as pipes are routinely inspected. Finally, the technology will enable funding models and assess the impact of different funding levels on the condition of infrastructure and repair backlog for water distribution. Successful completion of the project will include a multi-level technology tool that predicts condition over time with funding based on cross-sectional data for the DoD.

Currently, there are no existing tools within the DoD that specifically characterize linear assets, such as water distribution systems. Recent efforts have begun to characterize the current state of linear assets through a green, amber, and red categorical gradation. However, existing models to predict asset deterioration, like those used in BUILDER, are deterministic and require a continuous range of possible values from 0-100. Therefore, the tool will be an extension of the current categorical efforts for assessing the state of the infrastructure and directly enable forecasting and programming efforts for investment into water distribution models. The tool will reduce the need for continual inspection of assets, which is expensive as they are mostly underground, and will provide an informative tool to prioritize limited funding across the DoD and its installations. Direct outputs will validate assumptions for investment in infrastructure that largely rely upon investing 2% of total asset value in the infrastructure each year. The developed technology will critically evaluate this strategy and assess effectiveness of different investment strategies to improve overall condition of the assets and model repair backlog costs for advocacy of funding. (Anticipated Project Completion - 2025)