The iMin Workbench – A framework for predictive control with mineralogy

Since the introduction of automated mineralogy almost 40 years ago the holy grail has been to bring the capability to operational sites.  Having mineralogical information at our fingertips in a form simple enough to be used in day-to-day decision making is something that is well established in bringing huge benefits to operations, but it is only now, with the introduction of ruggedised SEM systems and cloud based expert support becoming a reality.  Pioneers such as Wolfgang Baum with Phelps Dodge and Robert Schouwstra with Anglo Platinum, among a swathe of others have highlighted the positives that integrating mineralogy into process optimisation can bring when applied on a project basis.  More recently the success that we have had at Kansanshi Copper Mine is testament that this can be extended to the day-to-day operation of large complex sites.  Now that this capability is at our fingertips it is time to start thinking about how all this extra data can be useful for operations, without overwhelming the site personnel and becoming more trouble than it is worth.

After the introduction of iMin Solutions I wanted to introduce the framework in which we are developing not only generation of mineralogy on-site but the support structures that need to be in place to make the most value from that investment.  Today I will introduce the iMin Workbench, which is the framework in which we will tie the generation of mineralogy information on-site with how it can be integrated with existing datasets to bring true predictive analytics to our operations.

How the iMin Workbench can be implemented to add revenue to your operation

The iMin Workbench takes a modular approach to developing an operation’s capability to use mineralogy as the glue that binds all the separate datasets we generate in a process circuit.  As the fundamental building block of what we are processing, mineralogy is the ultimate tool in relating process datasets that in isolation only look at very small parts of the circuit.  When combined using mineralogy these datasets provide a much more complete overview of what is driving process behaviour.  Once this link has been established the opportunity to predict process performance and reconcile against ore behaviour or geometallurgical models becomes a reality.

The iMin Workbench consists of three major modules:

– iMin(mineral)

The technology for on-site operational mineralogy that allows the people on site to generate the data they need.

– iMin(account)

The framework that allows the operation to extend their metallurgical balance to a mineralogical balance.

– iMin(analyse)

Making use of the trend based mineralogy data from operational mineralogy and supporting datasets to bring predictive analytics to mineral processing.

The modules are tightly integrated and give operations a structured path to make sense of what can be extremely complex data.  The first step on this path is to establish the iMin(mineral) program at an operation and begin generating daily mineralogy information that can be built up into mineralogical trends.  On its own this trend data can be powerful in identifying process improvement opportunities and significant improvements are nearly always made once it is applied.  Using daily mineralogy we are still being reactive in our approach to solving problems but the time for that reaction can be reduced from weeks or months to days, making the remedial action have much more impact.

The iMin(account) package takes the operation to the next level by adding the capability to integrate mineralogical accounting into the standard metallurgical accounting framework.  This begins to focus on the flow of minerals through a circuit and highlights the flow on effects of poor process conditions early in the circuit on the later recovery stages.

Finally, by relating the live mineralogical balance from iMin(account) to other datasets, such as process control, reagent schemes, online sensors etc we have all the data we need to get into Big Data analytics and turn our reactive approach to processing into predictive analytics.

Mineral processing plants are complex entities in which so many variables are in play that it often seems impossible to accurately track what is going on.  By taking a systematic approach to establishing the right data to start breaking down this complexity it is possible to get more predictive in how we approach processing.  Mineralogy provides the glue that holds all our information together in an operation and the iMin Workbench provides the framework in which that information can be properly applied.

The first step on this journey is to start building a database of operational mineralogy today.  Talk with iMin Solutions about how we can set this up for your site and establish a roadmap so that once sufficient data is available you can begin making the most of it.

If you have any questions feel free to contact us directly.  Also if you are interested in finding out more about process mineralogy make sure you download our digital book, which gives some more details on how to build mineralogical capability and start getting real value for your investment