As many know, cost overruns caused by unforeseen geological challenges are commonplace for large infrastructure projects. Average overruns between 20 and 50 percent are unfortunately typical in the industry. Although thorough ground investigations can reduce this risk, these geotechnical drillings and laboratory test are both expensive and time consuming.
In a recent peer-reviewed study, we’ve proven that the method used by EMerald Geomodelling – using artificial neural network (ANN) to integrate airborne geoscanning data and geotechnical data – can reduce this markedly.
.png)
Massive reduction in boreholes
More concretely, the study looks at the accuracy of top-of-bedrock models produced with this method and compares it to models one would make by traditional drilling campaigns.
In short, the study proves that EMerald's method can produce bedrock models that rival the accuracy of traditional drilling campaigns that use only geotechnical soundings.
Using ANN based bedrock interpretation, we’ve proven that one needs 2 - 3,5 times fewer geotechnical drillings in the early phases of a project compared to ground investigations using only borehole data. This equals a reduction of about 50 - 70 percent.
Rapid and thorough interpretation rather than polished geophysical models
Moreover, the study further reveals how the principles behind EMerald’s unique workflows break with common state-of-practice approaches, even cases where geophysical methods are employed. Traditionally, engineering geophysicists spend time and energy to produce the best geophysical result, then they attempt to indirectly relate to the engineering behavior of the ground. Often, experts manually interpret these geophysical results, making is a resource-demanding analysis that is hard to justify in cost-conscious projects. The results from both synthetic and real tests underline how EMerald’s more automated interface detection method is not only faster but is also robust and reliable.
The value of the results
CEO of EMerald Geomodelling and co-author of the study, Andi A. Pfaffhuber, believes the results are significant.
“Although we’ve been confident in the effectiveness of our solution after applying it in projects around the globe, it’s a big step for us to have it recognized in scientific peer review. It both shows the validity of our method and confirms the effectiveness of it. The extent of constructive review comments we got in the process, underlined the uniqueness of our approach”, states Pfaffhuber.
Interested in learning more?
If you want to learn more about the research, the full article - “Integrated bedrock model combining airborne geophysics and sparse drillings based on an artificial neural network” - is available online for free at ScienceDirect. Read it here!
You can also contact the EMerald team directly, at hello@emrld.no or +47 481 32 299 (NO)/+1 650 656-0174 (US)
.jpeg)
