Bridge Digital Twins Creation Using an Unorthodox Approach

The design procedure for a new bridge is standardised, and the methodology is clearly set by design standards. These are mainly computational static schemes, load influences, and behaviour of details such as joints, supports, etc. However, suppose we need to know the response of an existing bridge. In that case, experience shows that these procedures are often not sufficient to achieve consistency of numerical results (strain, frequencies, displacements, stresses, etc.) with the measured values; it is often necessary to modify static schemes and include various side effects that are otherwise not considered or neglected in the design. Getting a quality digital twin is a relatively difficult task. This paper describes some interesting practical examples of how better numerical models – digital twins of bridges – have been gradually built and debugged.