Risk and Agility in Design
In my last article, I introduced the topic of risk reduction through a variety of means. In this article, I continue the exploration of risks and how we might manage them appropriately in the context of mining infrastructure...
I welcome contributions on points for which you feel more should be stated, or for which you disagree. The dialogue is open.
Risk Agility
This past week I shared a Huffington Post article regarding agile risk management, in which it was suggested that “risk agility is the key attribute of the survivors”, and that “compliance…is not an adequate form of risk management for the twenty-first century”.
I thought it a fitting article because it tied nicely to a couple of the themes I had planned on discussing:
- complexity causing an amplification of risk, and
- a reliance on historical methods, data, and false-positives to reassure ourselves that we have effectively managed our risks.
This article suggested that we should be willing and able to monitor and manage risks according to how they arise, to make decisions as our risks shift and change over time. The author wrote,
“Aversion to risk is dangerous and implies being stuck in another era and held at a standstill by the inertia of fear.”
Given, the context of the article is that risk averse businesses will not survive.
But what of the things that need to stand the test of time, regardless of the changing conditions that they may be exposed to? What of those things, like infrastructure we might design and build, that are more permanent, and might be difficult or even impossible to make adaptations to, once they are complete?
How can risk agility be applied in that context?
Agility in Design
When it comes to managing risks, I agree that we must be agile over time – always monitoring and re-assessing the status of our risks, looking out for new things that may cause us concern, and responding to those things in a timely and appropriate fashion.
I would also acknowledge that for the facilities that I have worked upon, that this is exactly how ongoing design was managed. These were dams that I monitored, or managed the construction of - the types that included building annual raises, as the larger tailings facilities typically are.
For every new raise, the dams were re-assessed, designed, and major risks avoided, using all of the information available at that particular stage, including new information discovered throughout the year:
- incremental lift and ultimate height of the dam, including if changes had been made in regard to the life of the mine
- the volume and types of materials to contain behind
- basal soil and underlying geological conditions
- water table and pressures according to design specifications
- characteristics of all subsurface and in-dam materials according to testing performed on each, and assumed conditions in alignment with design
- ongoing climate conditions, projections, and storm assumptions
- factors of safety applied,
- and so on…
Examples include a zoned earth-fill dam with an almost vertical, low permeable core; and slurry tailings derived dams, primarily with a vertical rise over time, beaching out a large distance to keep ponds at bay.
These were heavily monitored for performance, and quality control on the construction significant – in terms of construction, testing always to ensure materials and densities met the specification of the design.
In terms of monitoring, watching for changes in piezometric response, movement in directions or over particular rates which were out of alignment with what was to be expected.
And always, especially, looking for new conditions – new risks – that needed to be addressed, and responding to them as rapidly as possible.
Adjustments Made
For one facility, I can think of several factors that arose which drastically impacted how we went about constructing, and which altered all future design stages. I won’t name them all, but will highlight a couple "challenges" we encountered along the way…
In one instance, while double checking subsurface conditions below the future footprint of the dam, we discovered a distinct segment of clay layers which had previously gone undetected. These were materials that were not present further upstream where the main core of the dam sat, but might otherwise have allowed slippage of large parts of the toe of the dam in later stages. The results?
- A lot more checking to see where the extents of these materials lay.
- The introduction of a massive excavation in the toe of the dam footprint.
- Installation of a very large shear key component within the design, and
- Additional weight added at the toe to buttress movements, from that year forward.
Another issue that arose at another time was encountering looser-than-appeared materials, and then exposure of bedrock, on one embankment tie-in at the dam centerline, a full year earlier than anticipated. This was discovered through late surface preparation within the construction season. For that year, it was too late to alter the design and because this was a valley which also received incoming water from freshet meltwaters, we couldn’t just put a hold on construction to change anything at the time.
The discovery caused some very quick thinking, some in-the-field adaptations and additional foundation preparation, and an expansion and widening of the dam core to the upstream. This was done to allow additional time before a final and more permanent resolution could be made, and to ensure containment would hold well until after the permanent fix.
What I can say about these in-the-moment, and significant changes to the best made plans, is that the memories these events bring back never go away. They are things you learn from - your mind expands to think of the unknowns for future projects - you see the big picture and begin to fathom how the things you do might impact others. And they bring a sense of accomplishment when it is all done.
But they also bring incredible stress, and concern that you've done the right things, and a sense that, if these challenges can be avoided, they should be!
Changes to Come?
Responding to changes in field and climate conditions, and unexpected monitoring results is not a new thing for construction, but with respect to the major risks that might be associated with these massive infrastructure – cannot be the only thing to rely upon to keep everyone safe. After all, when we are talking about containment dams, our response time to adapt design and construct something alternative is limited by the amount of materials, and the speed at which they need to be moved, to avoid disaster.
To reflect back on Dr. Andrew Robertson's points from my last article, we are increasing risks ~20 fold every 1/3 century – with bigger, higher, and more volumetric facilities. And we aren’t changing (drastically) our methodologies for design, or for construction.
We are basing designs on the results of geotechnical testing on the materials we use to construct with, tests that are run under assumed conditions to represent future reality. Those results are compared against theories of soil mechanics produced through extended testing of materials under a range of conditions. A reflection based on historical data, on what we might draw comfort in believing to be true.
But maybe we should consider that, over time, with changing climate conditions, or with the changes we are creating through the construction of larger infrastructure, we could be generating conditions that are beyond the understanding of current soil mechanics theory. There are many questions to consider:
Are we, over the construction and operative life of these facilities, changing the primary soil properties of the very materials that these facilities were built with?
Have we considered that under future conditions, the materials might behave differently than under which they were tested?
What are the limits of our current understanding of soil mechanics, and have we assumed the right parameters for use within our designs?
Should we be trying to learn more about these exceptional (long standing, high pressure) conditions – and perhaps looking to broader examples in nature again (more relevant to the applications we are now designing), to further study and validate our assumptions?
For infrastructure of such magnitude, of such high potential risk, what else can we be doing to identify future potential failure mechanisms - and eliminating them?
I'd love to hear your thoughts!