Visualization requirements of engineers for risk assessment of embankment dams

Abstract

Background: Aging infrastructure in the US has gained quite a bit of attention in the past decade. Being one type of a critical infrastructure, embankment dams in the US require significant investment to upgrade the deteriorated parts. Due to limited budgets, understanding the behavior of structures over time through risk assessment is essential to prioritize dams. During the risk assessment for embankment dams, engineers utilize current and historical data from the design, construction, and operation phases of these structures. The challenge is that during risk assessment, various engineers from different disciplines (e.g., geotechnical, hydraulics) come together, and how they would like to visualize the available datasets changes based on the discipline-specific analyses they need to perform. The objective of this research study is to understand the discipline-specific visualization needs of engineers from US Army Corps of Engineers (USACE) who are involved in risk assessment of embankment dams when they deal with large set of data accumulated since the inception of dams. Methods: The requirements were identified through a three-phased research approach including interviews with engineers who are regularly involved in risk assessment processes, a card game and review of standards and published work on risk assessment of embankment dams. Results: This paper provides the findings of research conducted with engineers coming from different disciplines within USACE. Findings comprise discipline-specific visualization requirements of engineers for viewing large datasets, containing static data (e.g. design information) and time-series data (e.g. piezometer data, monument measurements etc.), accumulated since the inception of dams. Conclusions: The findings suggest that the visualization of the dam layout, components and geometry within 3D settings overlaid with sensor data (which could be queried based on engineers’ discipline-specific needs) and data analytics results provide a better flexibility to engineers to understand the risk associated with potential failure modes.