Earlier this year we had the opportunity to sit down with Justin Glanvill and Nick Szebor from AMC Consultants to discuss how Datamine Studio RM assisted in overcoming their modelling and engineering challenges on-site. As a long-standing customer of Datamine, AMC trust in our software’s ability to implement data modelling and engineering methodologies that are highly specific and customised to individual mine sites.
“Few other packages out there give you the level of fundamental control over your source data that Datamine does, which is a very valuable tool within the consultancy space because we deal with such disparate datasets of such different quality.” Justin Glanvill.
“What people call the basic functions are very powerful – so the extra commands, the ability to tailor macros and scripts for the software. I think that’s where I find the best benefits of the software packages. That raw adjustability that we can get to the nuts and bolts of the data and really pour more out of it than possibly with other software.” Nick Szebor.
What challenges did they face?
One of the greatest challenges AMC constantly face are having to work with older datasets that aren’t quite as accurate or complete as newer data. In these cases, re-drilling and re-capturing data is often seen as the ideal solution to produce the most accurate and reliable geological models, however, in most cases this just isn’t economically and logistically viable. One example provided by Justin was “a client that had a large legacy dataset of geotechnical information, and they needed this information incorporated into a three-dimensional model to enhance their support planning and support requirements within the mine plan.” This is a very common problem in mines that have operated for a long period of time – they often only have incomplete or sparse historical data available.
Recently, AMC were able to help improve the design and scheduling of a large underground mine by constructing a reliable geotechnical model containing RQD and Q prime derived values, all from using sparse historical data in a database of hand-written logs that originally contained no direct RQD measurements.
How was this done?
Because the historical data did not originally contain direct RQD measurements, qualitative descriptions were instead used as a proxy to derive discrete indicator values to represent rock quality. A multiple indicator kriging methodology was used with the discrete values to estimate a model. Further post processing based on a probabilistic approach, including Monte Carlo simulations, were used to refine the MIK output resulting in a defensible and practical model that represented the deposit’s rock quality as accurately as possible.
What were the final benefits?
By using this method in Studio RM and incorporating legacy data into an RQD model, the need to re-gather new data through mapping and drilling was eliminated, which not only saved their clients a lot of time but also a lot of additional expenses. The model that AMC generated gave the engineers on-site a stable, robust platform to design the stopes and support requirements appropriately. Without these tools in Studio RM, the engineers would have had to work with lower resolution assumptions that had higher potentials for inaccuracy, potentially changing the economics of the stopes, planned extraction rates and extraction sequence.
By working with the toolset available in Studio RM, AMC were able to generate optimal results through time and cost saving efficiencies that benefited both short and long-term planning for their clients.
To view the full case study, please visit our webpage: