Digital Twins for Infrastructure

A digital twin is only useful while it stays synchronised with the physical asset. The engineering problem is therefore less about building the initial model and more about maintaining it over a service life measured in decades.

Update strategies

We compare three approaches: scheduled re-survey, event driven update after an inspection or repair, and continuous update from embedded sensors. Continuous update sounds attractive but produces a data volume that most asset owners are not prepared to store or interpret.

What to model, and what not to

The most common failure we see is over modelling. A twin built to answer a defined set of questions stays maintainable; a twin built to represent everything becomes obsolete within a year.