ABSTRACT
The continuous safety and performance of infrastructures is of particular importance for a modern society. Demands for resource efficiency in the construction sector will require significantly longer service lives for structures in the future. New digital maintenance strategies can make this goal achievable. To realize the full potential of such digital maintenance strategies, asset management systems need to incorporate a heterogeneous and linked set of data and information such as condition data, digital models of the geometry and damage characteristics. At the same time, all available data and information must be easily accessible for decision makers to evaluate the current state of the structures and derive maintenance actions from it. This paper presents a cloud-based asset management platform that incorporates digital twin models obtained from automated image-based inspection and damage detection methods, enabling predictive maintenance based on permanent storage and tracking of highly accurate information about the location and extent of damage.
