Reading: Summary, Note & Table Completion: Question 3

A digital twin is a continuously updated virtual model of a physical structure, built from a combination of the original design data and a live stream of readings from sensors embedded in the structure itself. Unlike a static 3D drawing produced once during construction, a digital twin changes in near real time: as a bridge deck expands in summer heat or a wind-turbine blade flexes under a gust, the corresponding change appears almost instantly in the virtual model, allowing engineers to monitor a structure's condition without physically inspecting it.

The approach was first applied at scale to critical transport infrastructure. In one widely cited case, engineers fitted a century-old steel railway bridge with several hundred strain gauges and accelerometers, instruments that measure, respectively, how much a material stretches under load and how strongly a structure vibrates. Data from these sensors is fed continuously into a computer model that recalculates the stress distributed across every rivet and girder of the bridge, updating the simulation roughly once every second. Engineers responsible for the bridge reported that the digital twin allowed them to identify a developing fatigue crack in a support beam several months before it would have become visible to a human inspector during a routine walkover survey.

Building an accurate digital twin depends heavily on machine learning. Because raw sensor data is noisy and enormous in volume, engineers train pattern-recognition models on years of historical readings so that the software learns to distinguish an ordinary, harmless fluctuation, such as the bridge deck's normal thermal expansion, from an early sign of structural fatigue. Over time, the trained model becomes more accurate at flagging genuine anomalies and produces fewer false alarms, reducing the number of unnecessary physical inspections that maintenance teams must carry out.

Cost is the main obstacle to wider adoption. Retrofitting an existing structure with a dense enough sensor network can cost several million pounds, a sum that is difficult to justify for smaller or less critical assets such as a rural footbridge. For this reason, digital twin technology has so far been concentrated on the highest-value infrastructure, including major road bridges, offshore wind turbines and large dams, where the cost of an unexpected structural failure vastly exceeds the price of continuous monitoring.

Complete the table below. Choose NO MORE THAN TWO WORDS from the passage for each answer.