INSIGHT | DATA, DIGITAL & BIM
Right, figure 4:
Integrating real-time surveying and
construction data into GIS for efficient progress tracking and coordination
while risk prediction models can integrate multiple
data streams to provide early warning of potential challenges. For a tunnel project in North America, an ML
algorithm is applied to an integrated dataset of geological conditions, derived abrasivity index, and earth pressure balance (EPB) TBM operational parameters — penetration rate, total thrust, chamber pressure, tool travel distance, and work done by the cutterhead to estimate tool wear rates. The uncertainty from geotechnical conditions generated an output of tool wear rates in terms of confidence intervals that captured the tool wear from the field observations.
Developing a Digital Archive The concept of a comprehensive digital archive is one of the most transformative applications of cohesive knowledge systems in tunnelling. By systematically capturing, organising, and linking all project information throughout the construction life cycle, these archives create novel opportunities for risk management and knowledge preservation that extend far beyond individual project boundaries. Traditional project documentation provides limited
insight into the complex relationships between design decisions, construction methods, geological conditions,
and project outcomes. Digital archives integrate these elements into comprehensive narratives that reveal relationships and patterns that would otherwise remain hidden. A good example application of a digital archive is
to understand and compare the design interpreted geotechnical conditions and those encountered during construction. When geotechnical conditions differ from initial investigations, the archive automatically links these discoveries with TBM performance adjustments, construction schedule impacts, and cost implications, creating a complete picture of how projects adapt to unexpected conditions. ML algorithms trained on comprehensive project
archives can identify subtle patterns in geotechnical data that indicate potential instability risks, capture TBM-ground interaction, and recommend optimal construction strategies based on successful approaches to similar challenges. This predictive capability becomes increasingly valuable as the archive grows and algorithms become more sophisticated. Digital archives enable statistical analysis of historical
risk data, improving quantification accuracy over expert judgment alone and supporting better project budgeting, insurance strategies, and client communication. This information proves invaluable for optimising future
Right, figure 5:
Workflow for information exchange during construction,
including an example of machine learning (ML) model output estimating tool wear
34 | October 2025
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45
