When the Challenger space shuttle exploded off the coast of Florida on January 28, 1986, Wayne Hale was head of the Propulsion Systems Section, Systems Division, Mission Operations, NASA. If you think you’ve experienced systems failure in your job, imagine if the entire event was being televised live around the world to hundreds of millions of people. Hale went on to become NASA Flight Director and Space Shuttle Program Manager and has recorded ten enduring lessons from his experience on how to avoid another Challenger-type incident. One of those lessons is that “a preoccupation with failure results in high reliability organizations.” He believes that dissension during the decision making process has tremendous value and that no dissension means the issue hasn’t been examined enough. Appoint devil’s advocates, he advises, and don’t let people remain silent – draw them out.
The world of underground coal mining, along with aviation, engineering, healthcare and computer security uses the Swiss cheese model of accident causation in risk analysis and risk management. The idea is that the risk of a threat becoming a reality is mitigated by the differing types of defenses which are “layered” behind each other – also known as the “cumulative act effect”. Unwanted events happen when the holes in our defences line up, “a trajectory of accident opportunity”, and there is no defence between us and the hazard.
Coal Mine Planning
As mining project managers, mine planning is, of course, one of our major defences, but I recently had cause to wonder whether our reliance on technology, or the poor use of it, was defeating our defences.
Is our growing reliance on technology actually making us complacent?
Are our systems giving us answers that seem precise but are, in fact, inaccurate?
Are we abdicating our decisions to “Swiss cheese” spreadsheets and models?
There are many objectives of a good coal mine plan, but an outcome we all want is an operation with acceptable (i.e. low) safety and operational risk combined with a high degree of reliability. Mine processes, after all, have significant technical and operational momentum that makes them time consuming and generally very costly to alter.
We tend to estimate the output from the mine by developing a production schedule based on the estimated capacity of the mining process measured in tonnes or metres cut per operating hour reduced by factors to account for the operational and resource risk.
If say the depth of cover increases, the forecast mining rate reduces. Other things also impact the mining rates such as weak roof or soft floor or geological structures such as faults or intrusions.
By putting these de-rating factors into the mine schedule we can forecast a risk adjusted production rate.
But here’s the rub – if we add a rule to the programme that says to only adjust the mining rate for the maximum de-rating factor encountered for a development road or longwall face, we may get an outcome that looks sound and is believable – but we may have missed a key message in the process.
In any location underground there may be one, several or many hazards present. If there is more than one hazard present, the risk is generally multiplied. Each risk on its own may be manageable from a mining perspective and the de-rating factor may be appropriate for that risk alone but if we combine a soft floor in a location with a heavy roof (cyclic loading being the culprit) and throw in a fault for good measure, rather than reduce the rate we may conclude that the area is unattractive and or unsafe for mining and best avoided.
The rule in the scheduling programme ignored the concentration of risk but if we checked by overlaying hazards on a plan or ran the schedule applying all the de-rating factors that are applicable at least as a first pass then the low rate of production that would have resulted for the at-risk location may have alerted the planners to a potential significant issue.
Risk management is more than spreadsheets and workshops and needs to be embedded in our coal mine management and planning processes.
Perhaps it has become too easy for us to point to the output of a spreadsheet without spending sufficient time making absolutely sure all of the necessary risks are not only contained within it but are being consolidated?
Have we become too comfortable with abdicating our decisions to the output of spreadsheets when they themselves might be a risk factor in our mine planning?
Bill Gates once wrote that “The ﬁrst rule of any technology used in a business is that automation applied to an efficient operation will magnify the efﬁciency. The second is that automation applied to an inefﬁcient operation will magnify the inefficiency.”
Tools are there to speed up processes but it pays to remain sceptical about the accuracy of the tools.
If we are to take Wayne Hale’s advice, dissension has tremendous value and we should all have someone inside our planning teams whose job it is to break apart our tools and look for systemic weaknesses, so we don’t have a Challenger experience.
Author: Rob Reeson, Partner, Enable Advisory – Mining Consultants Brisbane