Why the law of averages is the false premise behind every lost mining dollar

FORMER Gold Fields CEO, Ian Cockerill, was scathing at a recent mining conference regarding the reticence of the South African mining sector to embrace technology through the years. For too long, it had done its work “in the dark”, only discovering the results of its efforts as a result of long retrospection, and therefore unable to respond to problems with much effect. It was “a dinosaur,” he concluded.

That’s as damning as it comes.

To be fair to South African mining, however, much of its gold and some of its platinum mining is located deep underground where it is dark (and hot); and it is also highly labour intensive. Applying today’s leading edge technologies, which is to say digitalising the mining process, is therefore not the easiest of tasks; not when compared to open pit mining where access and visibility make surveillance that much easier.

Just ask Gold Fields. Try as it might, the company cannot seem to extract a decent margin from its Johannesburg West mine, South Deep. That is despite rounds and rounds of re-engineering initiatives, shift changes, and import of costly consultants.

In this regard, deep-level underground mining, for which South Africa is famed, is on the periphery of the much-vaunted ‘fourth industrial revolution’ in which companies like Business Science Corporation (BSC) are hoping to assist.

BSC – formerly Cyest Corporation – is a private technology company founded by Elton Bondi and Andreas Cambitsis, about two decades ago, which they claim has produced a profit from day one. One tends to believe that: Bondi comes across as a stickler for detail, whilst Cambitsis is – to caricature the two for a moment – the maverick beatnik.

(It might be also helpful to know that Bondi is a capped Protea pilot specialising in display aerobatics. That’s not like lawn bowls: one critical mistake and you used to be a capped Protea pilot specialising in display aerobatics.)

The area where BSC hopes to work with the mining industry is in the application of ‘digital twinning’. This is not new technology so much as new application of the concept. “It’s been around quite a long time,” said Bondi during a media presentation in BSC’s Sandton offices.

What’s different about it, however, is the role it plays in a convergence of different technologies: the so-called Internet of Things, in hi-fidelity visualisations, and in computational mathematics. The BSC acronym is there for a reason: Bondi & Co are boffins of note.

In essence, what BSC is hoping to do with digital twinning is show mining (and other) companies how to improve productivity by better understanding how the sum is driven by the parts. In its most granular manifestation, BSC can model in real-time how the operation and performance of, say, a mine’s yellow machinery, contributes towards EBITDA, or are not contributing to EBITDA, as the case may be.

In other words, when one says ‘twinning’, what’s actually happening is that a computational model, that can be displayed as a 3D diagram of a mine, is being recreated with the performance of its constituent parts laid bare.

It’s a micro-managers’ dream.

What’s interesting about the technology is that it refines time-honoured precepts about mining. One is the sacred cow of ‘volume and grade’, essentially a reference to the crucial aspects of profitability of any mine. Bondi also dismisses the notion of ‘average cost per tonne’. “There’s only value loss in the rule of averages,” said Bondi who thinks digital twinning can show that the ‘smoothing’ effects of average grade glosses over where pockets of value can be extracted.

It also shifts the debate somewhat away from what people mean by productivity which traditionally always revolves around mobilising a complement of employees.

There are limits to the technology though.

Where technology is badly needed in South African mining is in modifying human behaviour to improve safety. That’s hard for digital twinning to achieve as human beings are unpredictable, contrary and quite frequently act in ways that are unsafe. Some things can’t  be captured in a spread sheet.

1 COMMENT

  1. The last paragraph says it all.

    The control of behaviour, ultimately, is in the domain of the Regulator. Let’s wait until the post-election restructure of the DMR – specifically the Inspectorate – before we give blank cheques to all the Elon Musk wannabes.

LEAVE A REPLY