With seemingly endless options, identifying the right technology solution to provide value to given mine operation is a challenge. A new solution that forecasts the value, risk and business case of new mining technologies, could aid decision-making and help to speed up adoption by Australia’s mining industry.
Evaluating new technology for adoption
When assessing the future value of investment in a new technology, for example a fleet of autonomous load haul dump machines, several value and cost drivers emerge. These include increased productivity, improved safety and lower equipment maintenance costs.
Based on light historical data and assumptions, mining companies utilise traditional spreadsheet tools to establish a high-level business case for the new technology. These traditional spreadsheet tools tend to be large, uncontrolled and static, and are built with complex macros that only the developer and owner of the model truly understands.
A recent example in a 2013 report from the JPMorgan Chase Management Task Force, showed that reliance on the work of one “over-worked spreadsheet modeller” impacted the company in at least a AUD$400 million of financial losses in 2012.
A new model for technology assessments
Accurate technology assessments could help to speed up and increase adoption of new technologies by reducing risks to mining companies.
Research and development powerhouse, Mining3, has come up with a dynamic model that offers an alternative method for assessing technologies. Mining3 plans to eventually commercialise the technology to make it available to benefit the entire Australian mining industry.
"To ensure our model is effective, we firstly considered the uncertainties and variabilities of model inputs and the modelling approach used," Mining3 senior research engineer, Isaac Dzakpata, says.
"We needed to include comprehensive sensitivity and attribution analysis based on model input variabilities to gain an understanding of what will deliver best value for business."
Analyses covers not just the "most likely" scenario – but also the "best case" scenario (assuming the solution outperforms its targets). As well as the "worst case" scenario, where the solution grossly under-performs – we must understand any negative implications of the implementation of the technologies.
"Second, it is critical that the downstream and upstream effects of a solution are considered," Mr Dzakpata says.
"By effectively integrating the model with the appropriate lower level process levers, sensitivities of worst case downstream or upstream impacts can be adequately explored."
This means understanding capacity limits, bottlenecks and new opportunities that could be unlocked by an alternate solution.
A dynamic model of operational value chain
By developing a dynamic model that mimics the entire operational value chain, several solutions can be tested, independently or together, to arrive at an optimised system.
Mr Dzakpata says the third factor – and the most complex of the three – is the incorporation of risks into a model.
Whether this means understanding the risks of the new system failing or assessing the value of the risk that the new alternative system is mitigating.
"This step is often left out when it comes to developing a value model," he says.
"But, how can one properly assess the potential value of an autonomous truck system, if we haven't quantified the liability of a scenario in which an autonomous truck is involved in an accident?"
The outcome of this type of modelling approach would be a range of values based on a multi-factor decision criterion and not just a single central value expressed as a net present value, as is the case with most deterministic spreadsheet-based valuation models.
The challenge is understanding the economic and non-economic impact to the business as a whole, and assessing the impact dynamically. The ability to use a whole of business value modelling approach to understand issues like operational debottlenecking, life of mine opportunity trade-offs and change management implications is critical to whether the proposed value is overstated.
A whole-of-business approach
Mining3 has taken a whole-of-business, value driver tree modelling approach to assessing the new technology and methodologies. Utilising a top-down approach, they have developed a plug-and-play modular modeling structure, comprised of four main modules covering financial, production, task allocation and utilisation models.
In a hierarchical structure, in which the utilisation model (the lowest level set of inputs) flows into the highest level model (the financial model) that contains familiar financial metrics including cash flow, net present value and internal rate of return.
By building out the model from the top down, customisation is possible where a generic model can be applied to any mine and swiftly adapted for use on other mines by simply plugging in and plugging out modules.
Once a modular based model is built, data at the lowest level of the value drivers, such as equipment capacity, utilisation and commodity costs, will flow up to the high-level business case.
"By utilising ranges of inputs rather than static inputs, the capability of sensitivity and attribution analyses is unlocked, where small variances in a low-level input can be tracked through the model in a series of scenarios where the impact of the change of an input is quantified using stochastic or Monte Carlo style analyses," Mr Dzakpata says.
He adds that several challenges remain to be solved when it comes to value driver tree modelling of mining innovations.
"Firstly, the complexity of many mining task level relationships makes it difficult to build a dynamic model that effectively aggregates the task outputs into a total system output, and then replicates it for other processes.
"Second, while a model must be structured initially from the top level, the model and its modules must be adaptable such that it will allow future modelling of lower level inputs.
"Finally, successfully performing stochastic analyses on low-level inputs, while essential for the accuracy of technology assessment, is an onerous challenge requiring significant statistics expertise with a sound understanding of the interactions between low-level inputs that may alter the output values of the low-level inputs in a mining environment."
It's a big task and the Mining3 team, and its industry members, are working to complete the development of a one‑stop-shop, plug-and-play innovation value modelling tool that the mining industry can trust to assess new technologies and methodologies.
Significant strides have been made in developing this capability, but there is still work to be done before it will be available to the mining industry.
"The project is open for mining companies to get involved to help progress the whole-of-business value driver modelling system," Mr Dzakpata says.