NI 43-101 puts a fair amount of legal liability on the Qualified Person preparing a resource or reserve estimate or sign off on an advanced study. The QP is to be responsible for the accuracy of their work and take legal responsibility.
Every so often some new mining software comes along and I often wonder what are the risks in using it? Some examples of new mining software that I have heard about (but not personally used) nor have seen mentioned in any 43-101 studies are SimSched, the ThreeDify’s software packages, NPV One, and Bentley.
Is the software doing everything correctly?
Given that as a QP I am legally responsible for my work, I am bit apprehensive about how I can be assured the new software will provide reliable and accurate results for which I accept legal liability. The last thing I would want to do is issue a public technical report which is found to be in error due to a software bug.
Irrespective of 43-101, if you are working at a mining operation the last thing you want to do is present management with an incorrect reserve, pit design, or production plan.
If you are a consultant, how agreeable will your client be when you tell him that his study was done using a novel software package and not one of the industry standard packages, and there was an error in it?
I recall working with a major mining company and there was a reluctance to adopt any new software that was unproven and not an industry standard. Money was not the issue; the company’s concern was with the risk in using unproven software.
What if you have a limited budget?
How do you view new software if you have a limited budget? The new software may be cheaper, may appear to be be great, and may be a technological improvement and all at a lower cost. However the software risk still remains. There is no guarantee that all software output is correct simply because it comes from a computer.
As a QP, I suggest the onus is on the software developers to demonstrate that they can produce reliable and comparable results under all conditions. They need to be able to convince the future users that their software is accurate.
Perhaps over time the new software will gain wider adoption and be generally accepted. We may see more 43-101 reports that use it and hence it will get more overall acceptance.
Another question when developing a market for new software is whether it is better to focus on more consultant adoption or more mining company adoption?
Will mining companies use the software if their consultants are using it, or will consultants use it if more companies adopt it? It’s an interesting discussion that new software vendors must deal with in trying to grow their market share.
We are also faced with a plethora of “bigger and better” methodologies, all available at the touch of a button. Localized kriging, machine learning, implicit modeling, etc. While some are useful, a lot of these seem to be pushed by consulting firms advertising the next big thing. Nothing yet replaces time spent at the face, collecting data and making sure the basics are right, but more and more emphasis is being put on modeling in the office.
Good point Fred. Just because it is generated by a computer doesn’t mean its correct.